Creating an Investment Portfolio









1       STATEMENT OF THE PROBLEM

XYZ LLC business has grown from USD 200,000 in 2014 to USD 4,500,000 in 2017. This has been a phenomenal growth. As on March 2017, the turnover of XYZ India Pvt Ltd is Rs. 3,50,00,000. As a result, XYZ India Pvt Ltd has cash on hand which is lying idle in the Bank after meeting all its operational requirements. XYZ LLC would like to deploy this excess cash into Fixed Deposits, Mutual Funds and Shares. Doing so will ensure that the excess cash also earns income for the company.

However, as Mutual Funds and Share Market has risks in the investment made in it, XYZ LLC needs a strategy so that the deployed funds have a good chance of earning money which minimizing the involved risk. XYZ LLC requires an asset allocation strategy between Fixed Deposits, Mutual Funds and Shares such that maximum earning can be made from this portfolio at the very minimum risk.

The problem statement for XYZ LLC is to identify 5 Mutual Funds, between 6 to 10 Shares and up to 3 Fixed Deposit options to allocate a fund of Rs. 50,00,000 (Rupees Fifty Lakhs) such that this amount grows to a maximum possible amount within a time horizon of 5 years.

2       OBJECTIVE AND SCOPE OF THE STUDY

The objective of the study is to build a portfolio comprising of Shares, Mutual Funds and Fixed Income Instruments like Fixed Deposits, Recurring Deposits such that the returns are the maximum at the level of risk the Investor is ready to undertake. Shares and Mutual Funds of different categories constitute different levels of risk and returns. Also, the returns from different Fixed Deposits provide different levels of returns at different levels of risk. For example, returns from Fixed Deposits in Private Sector Banks give 6-7% per annum presently, while the same from Nationalised Banks provide 7-8% per annum, while that from Cooperative Banks provide 9-14% per annum. It needs to determine the asset allocation between the different instruments studying the risk-free rate and the present market rates.

The objective of this study is to identify 5 Mutual Funds, 6 to 10 Shares and up to 3 Fixed Deposit Options in which we could deploy Rs. 50,00,000 (Rupees Fifty Lakhs) such that the portfolio has the maximum chance of returning a good return within a time horizon of 5 years with the minimum possible risk.

The scope of the Study is to first determine an asset allocation strategy between Fixed Deposits, Mutual Funds and Shares for the total corpus of Rs. 50,00,000. For doing so, we will calculate the utility of the returns from the investment for the Investor as per the risk appetite of the Investor. 

We will consider 2 types of Fixed Deposits for study – Fixed Deposits in large Bank in India (We will consider this to be Risk Free Investment) and Fixed Deposits in Cooperative Banks (We will consider this to be a risky investment as is the case and this also provide much higher returns as compared to Fixed Deposits with Nationalized Bank or Large Private Bank in India).

After the asset allocation between Fixed Deposits, Mutual Funds and Shares (We will refer to these as Asset Classes), the scope of the study is to determine asset allocation within each of the Asset Class. In other words, once we have identified 5 Mutual Funds to invest in, we will determine what is the best allocation of funds allocated to the Mutual Fund Asset Class to each of the Mutual Funds so that the risk is minimised and the returns is maximised. The same will be done for Fixed Deposits and for Shares.

In the case of Fixed Deposits and Mutual Funds, it is possible to create a plan upfront for the next 5 years. However, in the case of Shares, it is not possible to plan upfront beyond a horizon of 1 year. So, this study will restrict to determining the shares to invest in the first year and will provide the different algorithms to apply for the subsequent years.

3       METHODOLOGY

There are 6 parts to this study. They are as follows:

1.   Allocation of funds to the different Asset Classes. The Asset Classes are Fixed Deposits in Large Indian Banks, Fixed Deposits in Cooperative Banks, Mutual Funds and Shares.

2.   Allocation of Funds allocated to the Fixed Deposit Asset Class between the different Fixed Deposits i.e. Fixed Deposits in Large Indian Banks and Fixed Deposits in Cooperative Banks.

3.   The strategy to be used for selection of Mutual Funds to be invested in.

4.   Allocations of Funds allocated to Mutual Fund Asset Class between the 5 Mutual Funds chosen for Investment.

5.   The strategy to be used for selection of Shares to be invested in.

6.   Allocation of Funds allocated to Shares Asset Class between 6 to 10 Shares chosen for Investment.

We will adopt different methodology for each of these 6 cases stated above. These 6 methodologies are stated below.

 

3.1    METHODOLOGY FOR ALLOCATION OF FUNDS BETWEEN ASSET CLASSES

We need to determine the portfolio in which funds are allocated between Fixed Deposits, Mutual Funds and Shares so that the Risk is minimum and the Returns are maximum.

3.1.1  RISK

We will measure Risk in terms of Standard Deviation of the Returns from the Asset Class.

We will consider investment in Fixed Deposits in Large Banks in India to be Risk-Free (or Zero Risk).

For Fixed Deposits in Cooperative Banks, we will gather the returns from these investments over the past 5 years. We will take the Standard Deviation of these Returns to determine the Risk in Investment in Fixed Deposits in Cooperative Banks.

For Mutual Funds, a Probabilistic Model will be created based on inputs from 3 Investment Bankers. The Probabilistic Model will state the probability of Returns in Mutual Funds in Good Times and Bad Times. Based on this, the Risk in Investment in Mutual Funds will be determined.

For Shares, the NIFTY Index will be gathered from 01-Jan-2017 to 31-May-2018. The Standard Deviation of the NIFTY Index over this period would be considered as the Risk in Share Investment.

3.1.2  RETURNS

For Fixed Deposits (both from Large Indian Banks and also from Cooperative Banks), we will consider the current rate of return as the rate of return.

For Mutual Funds and Shares, we will take inputs from 3 Investment Bankers regarding what is their projection of Returns from the Market in Good Times and Bad Times. Based on this, a Probabilistic Model will be created to determine the returns from Mutual Funds and Shares (separately).

3.1.3  ASSET ALLOCATION

As a strategy, to ensure that the portfolio is well diversified, we will ensure that we do not allocate more than 40% in any one Asset Class.

We will make an initial allocation between the Asset Classes. From this allocation, we will get the Expected Returns from the Portfolio as well as the Risk (Standard Deviation). From this, we will compute the Utility of the Investment for the Investor using the formula as follows:

Utility = E(R) – (1/2) * (A) * (V)

Where, E(R) is the Expected Returns, A is the Coefficient of Risk Aversion and V is the Variance.

Then, to distribute the corpus between the different Asset Classes, we will use Linear Programming so that the Utility from the Portfolio is maximum for the Risk Appetite of the Management.

To solve the Linear Programming problem, we will use Solver in Excel.

3.2    METHODOLOGY FOR ALLOCATION OF FUNDS WITHIN FIXED DEPOSITS ASSET CLASS

We will consider Fixed Deposit in Large Indian Banks to be Risk Free. 

We have seen that Reserve Bank of India (RBI) has been reducing the REPO Rate for last 4 years or have been keeping it unchanged. It is only after 4 consecutive years, that RBI has increased REPO Rate in June, 2018 by 0.25%. Experts feel that RBI could raise REPO Rate one more time before the next General Elections.

Seeing the current Indian Scenario, we will speculate that the current BJP Government will be voted to power in 2019. Under this assumption, we will consider that the current Government Policies will be continued. Under that circumstances, in all likelihood with the current Economic Indicators, there is very less possibilities for REPO Rate to be increased by RBI.

This assumption is important for our methodology for investing in Fixed Deposits as we will invest for a long term at the very beginning so that we do not face a situation for reduced Return Rate from Fixed Deposit in the future.

So, we will adopt the methodology that we will invest in Fixed Deposit for 5 years or more at the current rate of return. The current rate of return from Large Indian Bank can be obtained from their website. The current rate of return from Cooperative Bank will be determined by visiting the bank and getting this information. 

Further, as the company has account in HDFC Bank Ltd, we will consider investment in Fixed Deposit in Large Indian Bank to be in HDFC Bank Ltd. For the investment in Fixed Deposit in Cooperative Bank, we will consider investing in Ananda Cooperative Bank and Gnana Shale Souharda Cooperative Bank. The choice of Gnana Shale Souharda Cooperative Bank is because the company interacts with this Bank for all stamp paper requirements and thus can closely watch this Bank. As for Ananda Cooperative Bank, the office of this bank is right next to the offices of the company. Investing in 2 cooperative banks will reduce the risk as it will provide for diversification.

To determine the allocation, we will find the Standard Deviation of the Fixed Deposit Portfolio. We will consider this to be a 2-asset portfolio with one asset being Fixed Deposit in Large Indian Banks (Risk-Free) and one asset being Fixed Deposits in Cooperative Banks (Risky). 

We will consider the Correlation between the Fixed Deposit in Large Indian Bank and Fixed Deposit in Cooperative Bank to be 1. This is because the interest rate on both these types of Fixed Deposits will always move in the same direction whenever there is a REPO rate change from RBI.

We will solve the Linear Programming Problem for minimising the Standard Deviation (or Risk) to find the allocation in the 2 asset portfolio.

To solve the Linear Programming problem, we will use Solver in Excel.

3.3    METHODOLOGY FOR SELECTION OF MUTUAL FUNDS TO INVEST IN

One of the main objectives for selection of Mutual Funds for the portfolio is to ensure that we have a well-diversified set of Mutual Funds so that the risk is minimised. We will try to diversify the Mutual Fund portfolio from various points of view.

3.3.1  INDEX MUTUAL FUND

The first criteria for diversification will be between Funds are based on the Index and Funds that are not based on the Index (or are managed by Fund Managers). As our Investment horizon is 5 years, investing in both these types of Mutual Funds is important for us. It is generally observed that Index Based Mutual Funds outperform the Funds managed by Fund Managers over a period of 8 year. Research shows that there is only a 2% chance that a Mutual Fund managed by a Fund Manager will beat an Index based Mutual Fund after 8 years of investment.

As we have to select 5 Mutual Funds, we will keep one Mutual Fund as an Index Fund. We will study at least 2 Index Mutual Funds and select one out of this set.

To study the Index Mutual Funds, we will gather the Returns from the Index Mutual Funds over the last 5 years. From this data, we will compute the Average Returns as on date. This will be the value for Return on Asset from the Index Mutual Fund.

Also, from the returns data, we will be able to compute the Standard Deviation of the Index Mutual Fund. This will provide the component of risk involved in the Index Mutual Fund.

Using the value from Return on Asset and Standard Deviation, we can compute the Sharpe Ratio for the Index Mutual Fund using the following formula.

Sharpe Ratio = (Return on Asset – Risk Free Rate) / (Standard Deviation)

The Returns from Fixed Deposits in Large Indian Banks will be considered as the Risk Free Rate.

The Index Mutual Fund with the highest Sharpe Ratio will make it to our Portfolio of Mutual Funds.

3.3.2  MUTUAL FUND MANAGED BY FUND MANAGERS

With regards to selecting Mutual Funds managed by Fund Managers, we will create 2 types – one is Mutual Funds mostly invested in Equities and one is Mutual Funds mostly invested in Debt instruments.

For selecting the Mutual Fund mostly invested in Debt instruments, we will study the recommendations as available in Economic Times web site. We will select ONE Debt Fund which has the highest Sharpe Ratio.

So, now we will have to select 3 more Mutual Funds and these will be Mutual Funds mostly invested in Equities. For selecting the 3 Equity based Mutual Funds, we will study at least 5 Equity based Mutual Funds. We will compute the G-Score and F-Score for all these Mutual Funds. We will consider the top 50 percentiles of these Equity based Mutual Funds for further selection. From among the Equity based Mutual Funds in the top 50 percentiles, we will choose the set of 3 Mutual Funds which are most diversified as per the correlations between the returns of the Funds.

G-Score was formulated by Professor Mohandas. It is applicable for the Growth Shares.

Calculating G-Score

G-Score is an integer value between 0 and 8.

Companies with higher G-Scores are favourable for investment.

To calculate the G-Score, we need to consider the following 8 aspects of a Company and assign a score of 0 or 1 as per the criteria provided below for each aspect. Then we sum up the scores of all the 8 aspects to arrive at the G-Score for the Company.

The 8 aspects and the criteria are provided below.

Sl.

Aspect

How to calculate

Scoring Criteria

G1

Earnings return on assets

(Net income before extraordinary items) / (Average Total Assets for last 2 Years)

Assign 1 if value is greater than Industry Median; otherwise assign 0

G2

Cashflow return on assets

(Cash from Operations) / (Average Total Assets for last 2 Years)

Assign 1 if value is greater than Industry Median; otherwise assign 0

G3

Accruals

Gather figures for Cash from Operations (CFO) and Net Income (NI)

If CFO > NI, assign 1; otherwise 0

G4

Stability of earnings

Find Variance of Company’s quarterly earnings return on assets over the last 4 years

Assign 0 if value is greater than Industry Median; otherwise assign 1

G5

Sales Growth variability

Find Variance of a Company’s quarterly growth of sales over the last 4 years

Assign 0 if value is greater than Industry Median; otherwise assign 1

G6

R & D intensity

(Amount Spent on R&D in this Year) / (Total Assets at the beginning of the Year)

Assign 1 if value is greater than Industry Median; otherwise assign 0

G7

Capital expenditure intensity

(Amount Spent on Capital Expenditure in this Year) / (Total Assets at the beginning of the Year)

Assign 1 if value is greater than Industry Median; otherwise assign 0

G8

Advertising expense intensity

(Amount Spent on Advertising in this Year) / (Total Assets at the beginning of the Year)

Assign 1 if value is greater than Industry Median; otherwise assign 0

 

F-Score was formulated by Piotroski. It is applicable for all the Value Shares.

Calculating F-Score

F-Score is an integer value between 0 and 9.

Companies with higher F-Scores are favourable for investment.

To calculate the F-Score, we need to consider the following 9 aspects of a Company and assign a score of 0 or 1 as per the criteria provided below for each aspect. Then we sum up the scores of all the 9 aspects to arrive at the F-Score for the Company.

The 9 aspects and the criteria are provided below.

Sl.

Aspect

How to calculate

Scoring Criteria

F1

Return on Assets (ROA)

(Net income before extraordinary items) / (Total Assets at the beginning of the year)

Assign 1 if ROA is > 0; otherwise assign 0

F2

Cashflow from Operations (CFO)

(Cash from Operations) / (Total Assets at the beginning of the year)

Assign 1 if CFO > 0; otherwise assign 0

F3

DROA

Current Year’s ROA – Previous Years ROA

If DROA > 0, assign 1; otherwise 0

F4

Accrual

ROA = (Net income before extraordinary items) / (Total Assets at the beginning of the year)

CFO = (Cash from Operations) / (Total Assets at the beginning of the year)

Assign 1 if CFO > ROA; otherwise assign 0

F5

DLeverage

(Current Year Long-Term Debts / Average Total Assets for the last 2 years) – (Previous Year Long-Term Debts / Average Total Assets for the last 2 years)

Assign 0 if value is greater than 0; otherwise assign 1

F6

DLiquid

(Current Ratio for the Current Year) – (Current Ratio for the Previous Year)

Current Ratio = (Current Assets) / (Current Liabilities)

Assign 1 if value is greater 0; otherwise assign 0

F7

Equity Capital

(Amount of Equity offered in the Current Year)

Assign 1 if value is less than or equal to 0; otherwise assign 1

F8

DMargin

(Gross Margin Ratio for Current Year) – (Gross Margin Ratio for Previous Year)

Gross Margin Ratio = (Gross Margin) / (Total Sales)

Assign 1 if value is greater than 0; otherwise assign 0

F9

DTurnover

(Asset Turnover Ratio for Current Year) – (Asset Turnover Ratio for Previous Year)

Asset Turnover Ratio = (Total Sales) / (Total Assets at the beginning of the Year)

Assign 1 if value is greater than 0; otherwise assign 0

 

We will gather the information regarding which all shares the Equity Based Mutual Fund is invested in. From this information, we will gather which of the investments are investment in Growth Shares and which of these investments are in Value Shares. The Shares with high Price-to-Book value will be considered as Growth Shares. The Shares with low Price-to-Book value will be considered as Value Shares.

For all the Growth Shares, we will compute the G-Score.

For all the Value Shares, we will compute the F-Score.

We will take a weighted average of the G-Score and F-Score for an Equity Based Mutual Fund to arrive at the Score for the Mutual Fund.

Score for Mutual Fund = (%age Investment in Growth Shares * G-Score) + (%age Investment in Value Shares * F-Score) 

3.4    METHODOLOGY FOR ALLOCATION OF FUNDS WITHIN MUTUAL FUND ASSET CLASS

Once the Mutual Funds to be invested in have been identified, we will find the Returns from these Mutual Funds over the last 5 years or more. From this data, we will get the Average Returns from the Mutual funds. We will use the Average Returns over the last 5 or more years as the Return on Asset from the Mutual Fund.

Next, we will study the portfolio of each Mutual Fund. We will get the values of the weights in each Equity that the Mutual Fund has invested in. Also, for each of the Equity, we can get the value of b (Beta) for the Equity. Using the values of b for the individual Equities, we can compute the value of b for the Portfolio as the weighted sum of the individual b.

bPortfolio = w1 * b1 + w2 * b2 + … + wn * bn

where b1, b2, …, bn is the b for the individual equities and w1, w2, …, wn are the corresponding weights of allocations in these equities.

Using the values of Expected Returns and b, we can compute the Treynor Ratio for the Portfolio. Treynor Ratio can be computed using the formula as given below.

Treynor Ratio = (Return on Asset – Risk Free Rate) / b

The Returns from Fixed Deposits in Large Indian Banks will be considered as the Risk Free Rate.

We will formulate a Linear Programming Model for finding the Asset Allocation in each of the Mutual Funds such that the Treynor Ratio is maximised.

To solve the Linear Programming Model, we will use Solver in Excel.

3.5    METHODOLOGY FOR SELECTION OF SHARES TO INVEST IN

We cannot select stocks (shares) which we can stay invested for 5 years and expect returns. We will start with some purchases using a few strategies and then short (sell) and long (buy) based on various strategies.

For selecting Shares for investment, we will apply 3 different strategies. These strategies are as follows.

1.   Using SUE Score for selecting Shares for Investment

2.   Sloan’s Method for selecting Shares for Investment

3.   Pairs Trading Strategy

Each of the methods are explained below.

3.5.1  SUE SCORE METHOD

SUE is the acronym for Standardised Unexpected Earning.

SUE Score = (Actual EPS – Expected EPS) / (Standard Deviation of EPS)

·      EPS stands for Earning Per Share.

·      Actual EPS is the current EPS as released in the last year’s Financial Statements.

·      Expected EPS is calculated by finding the Average of the EPS at the end of the last 4 years (before the current year).

·      Standard Deviation of EPS is calculated by finding the Standard Deviation of the last 4 year’s EPS.

After finding the SUE Score of the Stocks of interest, order them in the descending order of the SUE Score.

As a thumb rule, we will sell all Stocks with negative SUE Score.

Also, we will consider selling all Stocks with very low SUE Score.

We will consider buying the Stock with the highest SUE Scores (Which are positive).

3.5.1.1              HOW LONG TO STAY INVESTED?

After investing in the Shares chosen through SUE Score, we will stay invested in the Shares for a minimum period of 1 year.

3.5.2  SLOAN’S METHOD

Richard G. Sloan had written a paper in 1996 for a trading strategy. This paper had shown that using this strategy profits were made for 28 out of 30 continuous years. This strategy is very useful for trading in the Indian markets today.

The strategy is based on the fact that we need to consider the amount of accruals in a company and not just the total revenue. The strategy states that companies with more amount of earnings from accruals will eventually generate lower profits for traders of the stock compared to companies which have higher cash equivalents and lower accruals.

The rationale of the strategy is that accruals may or may not convert into cash. This can happen because of accidents where the company fails to collect its dues from the services and products rendered. Also, it can be due to the company’s internal strategies to reflect better numbers to the traders.

One example of how accruals can be manufactured is from the automobile industry. The deliver managers generally push finished inventory to the distributors without firm orders. This immediately inflates the current receivables and thus increases the assets of the company. However, these are accruals and may not actually convert into cash. There is every likelihood that the Distributors may not be able to sell the inventory and return it to the manufacturer.

So, Sloan’s method is to find out the actual cash equivalent of a company and make decisions to buy or sell stocks based on that.

Sloan’s formula is calculated as follows:

Step 1: Calculate the Accruals (Let us call this “A”)

To calculate the accruals, the below formula needs using.

image001

Here,

·      DCurrent Assets = Current Assets of Current Year – Current Assets of Previous Year

·      DCash = Cash of Current Year – Current Cash of Previous Year

·      DCurrent Liabilities = Current Liabilities of Current Year – Current Liabilities of Previous Year

·      DTotal Short Term Debts = Total Short Term Debts of Current Year – Total Short Term Debts of Previous Year

·      DTax Payable = Tax Payable of Current Year – Tax Payable of Previous Year

·      Depreciation is taken for the Current Year.

Step 2: Determine the Total Income from Continuous Operations (Let us call this “I”)

This can be determined from the Income Statement of the Company. Some companies also call this “Statement of Profit and Loss”.

Step 3: Calculate Sloan Score = (I – A) / (Average Assets of Current Year and Previous Year)

The higher the Sloan Score, the Stock should be bought (or go long on the Stock). The lower the Sloan Score, the stock should be sold (or short it).

3.5.2.1              HOW LONG TO STAY INVESTED?

After investing in the Shares chosen through Sloan’s Method, we will stay invested in the Shares for a minimum period of 1 year.

3.5.3  PAIRS TRADING STRATEGY

“Pairs Strategy” is a short-term speculation Strategy. “Pair Trading” is in essence a Contrarian Investment Strategy. “Pairs Trading” is a medium term Trading Strategy.

3.5.3.1              The Strategy

 

1.   Find 2 Stocks whose prices have moved together for a long period of time.

2.   When the spread between them widens, short the Winner and long the Loser.

3.5.3.1.1       How to identify the Stocks that move together?

 

1.   Observe the Stocks Prices over a period of 12 months. This period of observation is called the FORMATION PERIOD.

2.   Construct a Cumulative Returns Index for each Stock over the Formation Period.

3.   Choose a Matching Partner for each Stock by finding a Stock which minimises the sum of squared deviations between the 2 Normalised Prices. (Normalised Prices can be found by using a method like taking each Stock Price and subtracting the Mean of the Stock Prices over the observation period and then diving this difference with the Standard Deviation of the Stock Prices over the observation period).

3.5.3.1.2       How far do the Stocks in the Pair have to Diverge before we can apply this Strategy?

When the Stock Prices of the Stocks in the Pair diverge by more than 2 Standard Deviations, then we start the TRADING PERIOD(The Trading Period has to be considered to be 6 month).

3.5.3.1.3       When should we unwind the position?

·      We should unwind the position at the point where the Stock Prices of the Pair crosses each other.

·      If the Stock Prices of Pair do not cross during the Trading Period, then we unwind on the last day of the Trading Period.

·      If one of the Stocks in the Pair is delisted, then we close position in the Pair.

3.5.3.2              Why this Strategy should earn returns?

By shorting the Winner and buying the Loser, if history repeats itself, the prices will converge and the arbitrageur will profit.

3.5.3.3              How to reduce Risk on the Portfolio through this Strategy?

The risk, after using the strategy, can be reduced by increasing the number of pairs selected for trading. As the number of Pairs in the Portfolio increases, the portfolio Standard Deviation falls. Also, it is noticed that as the number of Pairs in the Portfolio increases, the minimum realised returns increases; while the maximum realised excess returns remains relatively stable.

3.5.3.4              What is the observed returns from this Strategy?

For formulating this strategy, the researchers have studied data from the US Market between 1962 and 2002. They have found that adopting this strategy can yield 11% excess returns.

3.5.3.5              HOW LONG TO STAY INVESTED?

After investing in the Shares chosen through Pairs Strategy, we will stay invested in the Shares for a maximum period of 6 months.

3.6    METHODOLOGY FOR ALLOCATION OF FUNDS WITHIN SHARES ASSET CLASS

The amount allocated to Shares and to each individual Share will be dependent on the amount of risk the person is able to take. So, we will get a measure of the risk the investor is willing to take. The measure of risk in Shares can be defined by the b of the Share. Once we have the b for individual Shares, we can compute the b for the portfolio as follows.

bPortfolio = w1 * b1 + w2 * b2 + … + wn * bn

where b1, b2, …, bn is the b for the individual Share and w1, w2, …, wn are the corresponding weights of allocations in these Shares.

To allocate funds in each Share, we will form a Linear Programming Model such that the value of b is maximised within the upper limit for the b the investor is ready to undertake.

To solve the Linear Programming Model, we will use Solver in Excel.

 

 

 

4       SOURCE DATA

The source data is obtained from www.economictimes.com, www.bseindia.com, www.nseindia.com and www.moneycontrol.com.

Apart from this, we will scan the Internet for the Annual Reports of various companies. The list of the companies whose Annual Reports are studied are provided in Appendix I.

 

 

5       APPLYING MODEL FOR ASSET ALLOCATION BETWEEN ASSET CLASSES

We have to consider 3 Asset Classes – Fixed Deposits, Mutual Funds and Shares. To determine the allocation of funds between these Asset Classes, we first need to define the Returns and Risks in each Asset Class.

We define the Risk and Return for each Asset Class one by one.

We then define the correlation between the 3 asset classes.

Lastly, before the apply the model and allocate the funds between the asset classes, we will define Coefficient of Risk Aversion for the Investor.

5.1    RISK

5.1.1  FIXED DEPOSIT

5.1.1.1              INDIVIDUAL STANDARD DEVIATIONS

As stated earlier in Section 5.2, we will consider investment in Fixed deposit in for the following institutions.

1.   HDFC Bank Ltd.: This is Large Indian Bank. We will consider investment in Fixed Deposit in this Bank as Risk-Free. The Interest Rates provided by this bank over the last 5 years is as follows.

Year à

2018

2017

2016

2015

2014

Average Rate of Interest

7.10%

6.80%

6.50%

7.00%

7.90%

We see that there has been variation in Interest Rates over the last 5 years. However, we consider that this interest will be definitely returned by the Bank. Thus, we will consider the Risk in Investment in HDFC Bank Ltd as 0%.

We will denote HDFC Bank Ltd by H.

We will denote the Standard Deviation or Risk from Investment in Fixed Deposit in HDFC Bank Ltd as sH.

So, sH = 0.

2.   Ananda Cooperative Bank: This is Cooperative Bank. We will consider investment in Fixed Deposit in this Bank as Risky. The Interest Rates provided by this bank over the last 5 years is as follows.

Year à

2018

2017

2016

2015

2014

Average Rate of Interest

8.50%

9.00%

9.00%

9.50%

10.00%

The Standard Deviation of these Interest Rates is 0.57%. We will consider this to be risk in this Instrument. (Though it must be said that this is not an appropriate measure of this risk. However, we do not have any other measure at the moment)

We will denote Ananda Cooperative Bank by A.

We will denote the Standard Deviation or Risk from Investment in Fixed Deposit in Ananda Cooperative Bank as sA.

So, sA = 0.57%.

3.   Gnana Shale Souharda Cooperative Bank: This is Cooperative Bank. We will consider investment in Fixed Deposit in this Bank as Risky. The Interest Rates provided by this bank over the last 5 years is as follows.

Year à

2018

2017

2016

2015

2014

Average Rate of Interest

11.50%

13.00%

13.00%

13.50%

14.50%

The Standard Deviation of these Interest Rates is 1.08%. We will consider this to be risk in this Instrument. (Though it must be said that this is not an appropriate measure of this risk. However, we do not have any other measure at the moment)

We will denote Gnana Shale Souharda Cooperative Bank by G.

We will denote the Standard Deviation or Risk from Investment in Fixed Deposit in Gnana Shale Souharda Cooperative Bank as sG.

So, sG = 1.08%.

5.1.1.2              CORRELATIONS

From the above Interest Rates over the last 5 years, we can determine the correlations between these three Banks. Correlation between 2 instruments X and Y can be found using the formula as follows.

Correlation(X,Y) = Covariance(X,Y) / (sX * sY),

where sX is the Standard Deviation of X and sY is the Standard Deviation of Y.

So, the correlation table between the 3 banks is as follows.

 

H

A

G

H

1.00

 

 

A

0.53

1.00

 

G

0.41

0.78

1.00

So, Correlation(H, A) = rHA = 0.53

Correlation(H, G) = rHG = 0.41

Correlation(A, G) = rAG = 0.78

5.1.1.3              RISK FROM FIXED DEPOSITS

To find the risk from Fixed Deposits, we will initially assume that we will allocate equal amount into Fixed Deposits in HDFC Bank Ltd, Ananda Cooperative Bank and Gnana Shale Souharda Cooperative Bank. In other words, we will consider that the weights of investment in the Fixed Deposits in the 3 Banks wH, wA and wG are all equal to 33.33%.

With this assumption, we can find the Standard Deviation of the Fixed Deposit Portfolio (sP) as follows.

sP = wH2 * sH2 + wA2 * sA2 + wG2 * sG2 + 2 * rHA * wH * wA * sH * sA + 2 * rHG * wH * wG * sH * sG + 2 * wA * wG * rAG * sA * sG

Using this formula, we get the risk for the portfolio as follows.

sP = 2.73 * 10-5

So, we can consider the Risk from Investment in Fixed Deposits as 0.

5.1.2  MUTUAL FUNDS

For figuring out the Risk and Return from Mutual Funds, I consulted 3 Investment Bankers. They are :-

1.   Mr. Saket of Axis Securities Ltd. (C1)

2.   Mr. Rakesh VV of HDFC Securities Ltd. (C2)

3.   Mr. Sabyasachi Dash of Nirmal Bang Securities. (C3)

From them, I got the following figures regarding the possible returns from Mutual Funds.

Consultant

OPTIMISTIC RETURNS

PESSIMISTIC RETURNS

Probability

Return

Probability

Return

C1

70%

20%

30%

5%

C2

60%

22%

40%

5%

C3

60%

18%

40%

8%

From these figures, we can calculate the Returns as predicted by each of the Consultants.

 

Consultant

Calculation for Return

Return

C1

70% * 20% + 30% * 5%

15.50%

C2

60% * 22% + 40% * 5%

15.20%

C3

60% * 18% + 40% * 8%

14.00%

AVERAGE RETURN PREDICTED

14.90%

Now, we can calculate the Risk from the Mutual Funds as predicted by these 3 consultants. Note that Risk is measured as Standard Deviation. Standard Deviation is the SQUARE ROOT of Variance.

Consultant

Calculation for Variance

Risk

C1

70% * (20% – 15.5%)2 + 

30% * (5% – 15.5%)2

6.87%

C2

60% * (22% – 15.2%)2 + 

40% * (5% – 15.2%)2

8.33%

C3

60% * (18% – 14%)2 + 

40% * (8% – 14%)2

4.90%

AVERAGE RISK PREDICTED

6.70%

So, we will consider the Risk from Investment in Mutual Funds as 6.70%.

5.1.3  SHARES

For finding out the Risk involved in Shares, we will find the Standard Deviation of the NIFTY Index over the period between 01-Jan-2017 to 31-May-2018.

The NIFTY Index Data gathered is provided below.

 

2017

 

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

1

8,185.80 

8,716.40 

8,945.80 

9,173.75

9,304.05 

9,616.10 

9,520.90 

10,114.65 

9,974.40 

9,788.60 

10,440.50 

10,121.80 

2

8,179.50 

8,734.25 

8,899.75 

9,173.75 

9,313.80 

9,653.50 

9,520.90 

10,081.50 

9,974.40 

9,788.60 

10,423.80 

10,121.80 

3

8,192.25 

8,740.95 

8,897.55 

9,237.85 

9,311.95 

9,653.50 

9,615.00 

10,013.65 

9,974.40 

9,859.50 

10,452.50 

10,121.80 

4

8,190.50 

8,740.95 

8,897.55 

9,237.85 

9,359.90 

9,653.50 

9,613.30 

10,066.40 

9,912.85 

9,914.90 

10,452.50 

10,127.75 

5

8,273.80 

8,740.95 

8,897.55 

9,265.15 

9,285.30 

9,675.10 

9,637.60 

10,066.40 

9,952.20 

9,888.70 

10,452.50 

10,118.25 

6

8,243.80 

8,801.05 

8,963.45 

9,261.95 

9,285.30 

9,637.15 

9,674.55 

10,066.40 

9,916.20 

9,979.70 

10,451.80 

10,044.10 

7

8,243.80 

8,768.30 

8,946.90 

9,198.30 

9,285.30 

9,663.90 

9,665.80 

10,057.40 

9,929.90 

9,979.70 

10,350.15 

10,166.70 

8

8,243.80 

8,769.05 

8,924.30 

9,198.30 

9,314.05 

9,647.25 

9,665.80 

9,978.55 

9,934.80 

9,979.70 

10,303.15 

10,265.65 

9

8,236.05 

8,778.40 

8,927.00 

9,198.30 

9,316.85 

9,668.25 

9,665.80 

9,908.05 

9,934.80 

9,988.75 

10,308.95 

10,265.65 

10

8,288.60 

8,793.55 

8,934.55 

9,181.45 

9,407.30 

9,668.25 

9,771.05 

9,820.25 

9,934.80 

10,016.95 

10,321.75 

10,265.65 

11

8,380.65 

8,793.55 

8,934.55 

9,237.00 

9,422.40 

9,668.25 

9,786.05 

9,710.80 

10,006.05 

9,984.80 

10,321.75 

10,322.25 

12

8,407.20 

8,793.55 

8,934.55 

9,203.45 

9,400.90 

9,616.40 

9,816.10 

9,710.80 

10,093.05 

10,096.40 

10,321.75 

10,240.15 

13

8,400.35 

8,805.05 

8,934.55 

9,150.80 

9,400.90 

9,606.90 

9,891.70 

9,710.80 

10,079.30 

10,167.45 

10,224.95 

10,192.95 

14

8,400.35 

8,792.30 

9,087.00 

9,150.80 

9,400.90 

9,618.15 

9,886.35 

9,794.15 

10,086.60 

10,167.45 

10,186.60 

10,252.10 

15

8,400.35 

8,724.70 

9,084.80 

9,150.80 

9,445.40 

9,578.05 

9,886.35 

9,794.15 

10,085.40 

10,167.45 

10,118.05 

10,333.25 

16

8,412.80 

8,778.00 

9,153.70 

9,150.80 

9,512.25 

9,588.05 

9,886.35 

9,897.30 

10,085.40 

10,230.85 

10,214.75 

10,333.25 

17

8,398.00 

8,821.70 

9,160.05 

9,139.30 

9,525.75 

9,588.05 

9,915.95 

9,904.15 

10,085.40 

10,234.45 

10,283.60 

10,333.25 

18

8,417.00 

8,821.70 

9,160.05 

9,105.15 

9,429.45 

9,588.05 

9,827.15 

9,837.40 

10,153.10 

10,210.85 

10,283.60 

10,388.75 

19

8,435.10 

8,821.70 

9,160.05 

9,103.50 

9,427.90 

9,657.55 

9,899.60 

9,837.40 

10,147.55 

10,146.55 

10,283.60 

10,463.20 

20

8,349.35 

8,879.20 

9,126.85 

9,136.40 

9,427.90 

9,653.50 

9,873.30 

9,837.40 

10,141.15 

10,146.55 

10,298.75 

10,444.20 

21

8,349.35 

8,907.85 

9,121.50 

9,119.40 

9,427.90 

9,633.60 

9,915.25 

9,754.35 

10,121.90 

10,146.55 

10,326.90 

10,440.30 

22

8,349.35 

8,926.90 

9,030.45 

9,119.40 

9,438.25 

9,630.00 

9,915.25 

9,765.55 

9,964.40 

10,146.55 

10,342.30 

10,493.00 

23

8,391.50 

8,939.50 

9,086.30 

9,119.40 

9,386.15 

9,574.95 

9,915.25 

9,852.50 

9,964.40 

10,184.85 

10,348.75 

10,493.00 

24

8,475.80 

8,939.50 

9,108.00 

9,217.95 

9,360.55 

9,574.95 

9,966.40 

9,857.05 

9,964.40 

10,207.70 

10,389.70 

10,493.00 

25

8,602.75 

8,939.50 

9,108.00 

9,306.60 

9,509.75 

9,574.95 

9,964.55 

9,857.05 

9,872.60 

10,295.35 

10,389.70 

10,493.00 

26

8,602.75 

8,939.50 

9,108.00 

9,351.85 

9,595.10 

9,574.95 

10,020.65 

9,857.05 

9,871.50 

10,343.80 

10,389.70 

10,531.50 

27

8,641.25 

8,896.70 

9,045.20 

9,342.15 

9,595.10 

9,511.40 

10,020.55 

9,857.05 

9,735.75 

10,323.05 

10,399.55 

10,490.75 

28

8,641.25 

8,879.60 

9,100.80 

9,304.05 

9,595.10 

9,491.25 

10,014.50 

9,912.80 

9,768.95 

10,323.05 

10,370.25 

10,477.90 

29

8,641.25 

 

9,143.80 

9,304.05 

9,604.90 

9,504.10 

10,014.50 

9,796.05 

9,788.60 

10,323.05 

10,361.30 

10,530.70 

30

8,632.75 

 

9,173.75 

9,304.05 

9,624.55 

9,520.90 

10,014.50 

9,884.40 

9,788.60 

10,363.65 

10,226.55 

10,530.70 

31

8,561.30 

 

9,173.75 

 

9,621.25 

 

10,077.10 

9,917.90 

 

10,335.30 

 

10,530.70 

 

 

2018

 

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

1

10,435.55 

11,019.35 

10,458.35 

10,113.70 

10,739.35 

 

 

 

 

 

 

 

2

10,442.20 

10,760.60 

10,458.35 

10,211.80 

10,716.95 

 

 

 

 

 

 

 

3

10,443.20 

10,760.60 

10,458.35 

10,243.05 

10,668.50 

 

 

 

 

 

 

 

4

10,504.80 

10,760.60 

10,458.35 

10,120.20 

10,618.25 

 

 

 

 

 

 

 

5

10,558.85 

10,666.55 

10,358.85 

10,328.95 

10,618.25 

 

 

 

 

 

 

 

6

10,558.85 

10,498.25 

10,221.20 

10,331.60 

10,618.25 

 

 

 

 

 

 

 

7

10,558.85 

10,476.70 

10,163.55 

10,331.60 

10,715.50 

 

 

 

 

 

 

 

8

10,623.60 

10,576.85 

10,242.65 

10,331.60 

10,717.80 

 

 

 

 

 

 

 

9

10,637.00 

10,454.95 

10,226.85 

10,379.35 

10,741.70 

 

 

 

 

 

 

 

10

10,632.20 

10,454.95 

10,226.85 

10,401.05 

10,715.55 

 

 

 

 

 

 

 

11

10,651.20 

10,454.95 

10,226.85 

10,417.15 

10,806.50 

 

 

 

 

 

 

 

12

10,681.25 

10,533.70 

10,422.70 

10,458.65 

10,806.50 

 

 

 

 

 

 

 

13

10,681.25 

10,539.75 

10,426.85 

10,480.60 

10,806.50 

 

 

 

 

 

 

 

14

10,681.25 

10,500.90 

10,410.90 

10,480.60 

10,806.60 

 

 

 

 

 

 

 

15

10,761.50 

10,545.50 

10,360.15 

10,480.60 

10,801.85 

 

 

 

 

 

 

 

16

10,700.45 

10,452.30 

10,195.15 

10,528.35 

10,741.10 

 

 

 

 

 

 

 

17

10,788.55 

10,452.30 

10,195.15 

10,548.70 

10,682.70 

 

 

 

 

 

 

 

18

10,817.00 

10,452.30 

10,195.15 

10,526.20 

10,599.50 

 

 

 

 

 

 

 

19

10,894.70 

10,378.40 

10,094.20 

10,565.30 

10,596.40 

 

 

 

 

 

 

 

20

10,894.70 

10,360.40 

10,136.55 

10,564.05 

10,596.40 

 

 

 

 

 

 

 

21

10,894.70 

10,397.49 

10,155.25 

10,564.05 

10,521.05 

 

 

 

 

 

 

 

22

10,966.20 

10,382.70 

10,155.25 

10,564.05 

10,536.70 

 

 

 

 

 

 

 

23

11,069.35 

10,491.05 

9,998.05 

10,584.70 

10,429.05 

 

 

 

 

 

 

 

24

11,086.00 

10,491.05 

9,998.05 

10,617.75 

10,528.80 

 

 

 

 

 

 

 

25

11,069.65 

10,491.05 

9,998.05 

10,570.55 

10,605.15 

 

 

 

 

 

 

 

26

11,069.65 

10,582.60 

10,130.65 

10,617.80 

10,605.15 

 

 

 

 

 

 

 

27

11,069.65 

10,554.30 

10,183.70 

10,692.30 

10,605.15 

 

 

 

 

 

 

 

28

11,069.65 

10,492.85 

10,121.25 

10,692.30 

10,688.65 

 

 

 

 

 

 

 

29

11,130.40 

 

10,113.70 

10,692.30 

10,633.30 

 

 

 

 

 

 

 

30

11,049.65 

 

10,113.70 

10,739.35 

10,614.35 

 

 

 

 

 

 

 

31

11,027.70 

 

10,113.70 

 

10,736.15 

 

 

 

 

 

 

 

The Standard Deviation of these values is 680.70.

We scale this value by the minimum recorded value in this period.

The minimum recorded value is 8,179.50.

So, the Scaled Standard Deviation is 8.32%. We will consider this as the Risk in Investment in Shares.

 

5.2    RETURN

5.2.1  FIXED DEPOSIT

The returns gathered for the 3 banks is as follows.

Year

2018

2017

2016

2015

2014

HDFC Bank Ltd

7.10%

6.80%

6.50%

7.00%

7.90%

Ananda Cooperative Bank

8.50%

9.00%

9.00%

9.50%

10.00%

Gnana Shale Souharda Cooperative Bank

11.50%

13.00%

13.00%

13.50%

14.50%

We will take the average of the current Interest Rate as the Returns from Investment in Fixed Deposits.

Average of Current Returns = (7.10%+8.50%+11.50%)/3 = 9.03%

So, we will target 10% returns from Investment in Fixed Deposits.

5.2.2  MUTUAL FUND

For figuring out the Risk and Return from Mutual Funds, I consulted 3 Investment Bankers. They are :-

1.   Mr. Saket of Axis Securities Ltd. (C1)

2.   Mr. Rakesh VV of HDFC Securities Ltd. (C2)

3.   Mr. Sabyasachi Dash of Nirmal Bang Securities. (C3)

From them, I got the following figures regarding the possible returns from Mutual Funds.

Consultant

OPTIMISTIC RETURNS

PESSIMISTIC RETURNS

Probability

Return

Probability

Return

C1

70%

20%

30%

5%

C2

60%

22%

40%

5%

C3

60%

18%

40%

8%

From these figures, we can calculate the Returns as predicted by each of the Consultants.

Consultant

Calculation for Return

Return

C1

70% * 20% + 30% * 5%

15.50%

C2

60% * 22% + 40% * 5%

15.20%

C3

60% * 18% + 40% * 8%

14.00%

AVERAGE RETURN PREDICTED

14.90%

So, we will target 15% returns from Investment in Mutual Funds.

5.2.3  SHARES

For figuring out the Risk and Return from Shares, I consulted 3 Investment Bankers. They are :-

1.   Mr. Saket of Axis Securities Ltd. (C1)

2.   Mr. Rakesh VV of HDFC Securities Ltd. (C2)

3.   Mr. Sabyasachi Dash of Nirmal Bang Securities. (C3)

From them, I got the following figures regarding the possible returns from Shares.

 

C1

C2

C3

Probability

20%

30%

30%

Returns

60%

50%

30%

Probability

70%

50%

40%

Returns

20%

20%

15%

Probability

10%

20%

30%

Returns

-10%

-20%

-15%

Combined Return Formula

(20%*60%) +

(70%*20%) +

(10%*-10%)

(30%*50%) +

(50%*20%) +

(20%*-20%)

(30%*30%) +

(40%*15) +

(30%*-15%)

Returns

25%

21%

10.5%

Instead of taking average in this case, I will take the prediction of Mr. Saket of Axis Securities. This is because I have found that his advice has resulted in best income from Shares for me in the past.

So, we will consider returns from Investment in Shares as 25%.

5.3    CORRELATION BETWEEN ASSET CLASSES

I could not find reliable data regarding the correlation between Investment in Fixed Deposits, Mutual Funds and Shares. So, we will consider empirical values for the correlations based on economic intuitions.

It is noticed that Fixed Deposits and Mutual Funds are not very highly correlated. However, still they are positively correlated. This is because when the Economy does well, generally Fixed Deposit Rates falls as REPO Rate is slashed; while returns from Mutual Funds booms. The opposite is the case when Economy does not do well. 

However, Fixed Deposits and Mutual Funds give positive returns in good times.

So, we will consider correlation between Fixed Deposits and Mutual Funds to be 0.3.

By the same intuition, we can argue correlation between Fixed Deposits and Shares. So, we will consider correlation between Fixed Deposits and Shares as 0.1. We consider a lower value here as Mutual Funds are insulated by Debt Instruments during a downturn in Equity Markets.

Mutual Funds and Shares generally move together. However, as Mutual Funds are also invested in Debt Instruments and Debt Instruments move in opposite direction to Equities, we will consider correlation between Mutual Funds and Shares to be 0.6.

So, the correlation matrix is as follows:

 

Fixed Deposit

Mutual Fund

Shares

Fixed Deposit

1.0

 

 

Mutual Fund

0.3

1.0

 

Shares

0.1

0.6

1.0

 

5.4    COEFFICIENT OF RISK AVERSION

There are 3 types of Investors. They are as follows.

1.   Risk Averse: These Investors avoid taking Risk. The Coefficient of Risk Aversion for these Investors is >0.

2.   Risk Neutral: These Investors are neutral to taking Risk. The Coefficient of Risk Aversion for these Investors is =0.

3.   Risk Seekers: These Investors are willing to take Risk. The Coefficient of Risk Aversion for these Investors is <0.

As our Investor is an Entrepreneur and through Interview we gather that our Investor is a Risk Seeker. Empirically, we assign a value of -2 to the Coefficient of Risk Aversion for our Investor.

5.5    ALLOCATING FUNDS BETWEEN ASSET CLASSES

Now that we have all the building blocks, we allocate funds to the 3 Asset Classes using a Linear Programming Model to maximise the Utility from the Investment for the Investor.

The initial state of the formulation looks as follows.

image002

The setup of the Linear Programming Model is as follows.

image003

Now, we solve the Linear Programming problem to maximise the Utility. The results are as follows.

image004

So, the optimum Fund Allocation is as follows.

Instrument

Allocated Amount (Rs.)

Fixed Deposits

Rs. 10,00,000.00

Mutual Funds

Rs. 20,00,000.00

Shares

Rs. 20,00,000.00

 

We notice that the Expected Returns is 18% from this portfolio.

Also, notice that the Risk from this Portfolio is almost eliminated.

6       APPLYING MODEL FOR FUND ALLOCATION WITHIN EACH ASSET CLASSES

Now that we have determined how much funds to allocate to each Asset Class, we proceed to select Assets within each Asset Class and allocate Funds to them.

6.1    FIXED DEPOSITS

We have already strategized regarding which Fixed Deposits to invest in. So, now we need allocating funds to each Fixed Deposit such that the Risk is minimised.

For this, we set up the Linear Programming Model by initially allocating equal amount of funds to each of the 3 Fixed Deposits. The initial set up is as follows.

image005

The Linear Programming Model set up in Excel is as follows.

image006

On solving for this Linear Programming Model, we get the following results.

image007

So, we make the final allocation as follows:

Institution

Amount Allocated

HDFC Bank Ltd

Rs. 2,60,000.00

Ananda Cooperative Bank

Rs. 1,10,000.00

Gnana Shale Souharda Cooperative Bank

Rs. 6,30,000.00

image008

We notice that the Returns are still the same. However, Risk has increased; though it is negligible.

6.2    MUTUAL FUNDS

We first need to select the Mutual Funds to invest in. Once we have selected the Mutual Funds to invest in, we need to allocate funds to the same.

6.2.1  SELECTING MUTUAL FUNDS

We need to select primarily 2 types of Mutual Funds – Index based Mutual Funds and Non-Index based Mutual Funds (i.e. Mutual Funds managed by Fund Managers).

6.2.1.1              SELECTING INDEX BASED MUTUAL FUNDS

We will study the following 4 Index based Mutual Funds.

1.   Franklin India Index Fund NSE NIFTY

2.   HDFC Index Fund – NIFTY Plan

3.   ICICI Prudential NIFTY Index Fund

4.   SBI NIFTY Index Fund

The data gathered from www.economictimes.com are as follows.

FUND

RETURNS

1 year

3 years

5 years

Franklin India Index Fund

12.87%

10.52%

13.31%

HDFC Index Fund

14.21%

11.38%

14.16%

ICICI Prudential Index Fund

13.19%

10.83%

13.90%

SBI Index Fund

13.58%

10.89%

13.20%

Now, we calculate the Expected Returns (Er), Standard Deviation (s) and then the Sharpe Ratio.

FUND

Er

s

Sharpe Ratio

Franklin India Index Fund

12.23%

1.50%

3.42220

HDFC Index Fund

13.25%

1.62%

3.79709

ICICI Prudential Index Fund

12.64%

1.61%

3.44698

SBI Index Fund

12.56%

1.46%

3.74816

So, we select HDFC Index Fund – NIFTY Plan.

6.2.1.2              SELECTING DEBT BASED MUTUAL FUNDS

We will study the following 3 Debt based Mutual Funds. These 3 are recommended by www.moneycontrol.com. 

1.   IDFC Bond Fund

2.   Reliance Income Fund

3.   SBU Dynamic Bond Fund

The data gathered from www.economictimes.com are as follows.

FUND

RETURNS

1 year

3 years

5 years

IDFC Bond Fund

8.73%

7.77%

7.77%

Reliance Income Fund

6.95%

8.59%

8.59%

SBI Dynamic Bond Fund

8.82%

8.89%

8.89%

Now, we calculate the Expected Returns (Er), Standard Deviation (s) and then the Sharpe Ratio.

FUND

Er

s

Sharpe Ratio

IDFC Bond Fund

8.09%

0.55%

1.78618

Reliance Income Fund

8.04%

0.95%

0.99628

SBI Dynamic Bond Fund

8.87%

0.04%

43.71366

So, we select SBI Dynamic Bond Fund.

6.2.1.3              SELECTING EQUITY BASED MUTUAL FUNDS

We will study the following 6 Equity based Mutual Funds. These 6 are recommended by Axis Securities. 

1.   Axis Focused 25 Fund

2.   Axis Long Term Equity Fund

3.   Aditya Birla SL Frontline Equity Fund

4.   Franklin India Bluechip Fund

5.   IDFC Core Equity Fund

6.   Reliance Small Cap Fund

6.2.1.3.1       AXIS FOCUSED 25 FUND

The allocation of funds is as follows:

Equity

91.24%

Rs. 1422.13 crores

Debt

9.21%

Rs. 342.07 crores

Others

-0.45%

– Rs. 16.71 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

b

Kotak Mahindra Bank Ltd

7.85%

291.55

0.76

HDFC Bank Ltd

7.59%

281.74

0.69

Tata Consultancy Services Ltd

6.70%

248.95

0.25

Maruti Suzuki India Ltd

5.90%

219.20

0.92

Supreme Industries Ltd

5.88%

218.31

0.70

Shree Cement Ltd

5.58%

207.31

0.91

Bajaj Finance Ltd

5.01%

186.04

1.29

Bajaj Finserv Ltd

4.87%

180.93

1.33

Gruh Finance Ltd

4.36%

162.09

0.73

Motherson Sumi Systems Ltd

3.68%

136.79

1.29

 

We separate the Growth Stocks from the Value Stocks. For this, we find the Price-to-Book Value (P/B). A low P/B is considered as a Value Stock (V) and a high P/B is considered as a Growth Stock (G).

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

Gruh Finance Ltd

37.76

681.80

18.06

G

Bajaj Finance Ltd

174.59

2,100.60

12.03

G

Supreme Industries Ltd

133.49

1,264.80

9.47

G

Shree Cement Ltd

2,209.71

16,102.00

7.29

G

Maruti Suzuki India Ltd

1,227.84

8,810.40

7.18

G

Kotak Mahindra Bank Ltd

211.66

1,304.70

6.16

G

Bajaj Finserv Ltd

1,014.90

5,770.95

5.69

G

HDFC Bank Ltd

423.71

2,063.60

4.87

G

Motherson Sumi Systems Ltd

74.85

307.65

4.11

G

Tata Consultancy Services Ltd

446.80

1,721.60

3.85

G

 

For all the Growth Stocks, we need finding the G-Score. And for all the Value Stocks, we need finding the F-Score.

Given here are the G-Score Calculations for the 10 companies. (We do not include the complete calculation of G-Score for each company as this will occupy a lot of space in this paper)

 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

Kotak Mahindra Bank Ltd

1

1

0

1

1

0

1

1

6

5.88%

0.4710

Tata Consultancy Services Ltd

1

1

1

1

1

1

1

0

7

7.59%

0.4690

HDFC Bank Ltd

1

1

0

1

1

0

1

1

6

5.01%

0.4554

Maruti Suzuki India Ltd

1

1

0

1

1

1

1

1

7

7.85%

0.4130

Shree Cement Ltd

1

1

0

1

1

1

1

0

6

5.90%

0.3348

Bajaj Finance Ltd

1

1

1

1

1

0

0

1

6

6.70%

0.3006

Supreme Industries Ltd

1

1

0

1

1

0

1

0

5

4.36%

0.2940

Bajaj Finserv Ltd

1

1

1

1

1

0

0

1

6

5.58%

0.2922

Motherson Sumi Systems Ltd

1

1

0

1

1

1

1

0

6

4.87%

0.2208

Gruh Finance Ltd

1

1

0

1

1

0

0

0

4

3.68%

0.1744

NET G-SCORE

3.4252

There are no Value Stocks and thus we do not calculate the F-Score. 

So, the total score of this Fund is 3.4252.

6.2.1.3.2       AXIS LONG TERM EQUITY FUND

The allocation of funds is as follows:

Equity

95.20%

Rs. 16703.95 crores

Debt

2.87%

Rs. 504.11 crores

Others

1.93%

Rs. 338.64 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

HDFC Bank Ltd

7.78%

1,364.94

Tata Consultancy Services Ltd

7.65%

1,341.78

Kotak Mahindra Bank Ltd

7.62%

1,337.44

Pidilite Industries Ltd

6.41%

1,125.06

HDFC Ltd

6.05%

1,061.70

Bajaj Finance Ltd

5.22%

915.47

Gruh Finance Ltd

5.03%

883.09

Maruti Suzuki India Ltd

4.51%

791.70

Avenue Supermarts Ltd

4.33%

760.34

Motherson Sumi Systems Ltd

3.34%

586.45

 

We separate the Growth Stocks from the Value Stocks.

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

Avenue Supermarts Ltd

61.56

1,523.65

24.71

G

Gruh Finance Ltd

37.76

681.80

18.06

G

Pidilite Industries Ltd

70.18

1,078.05

15.33

G

Bajaj Finance Ltd

174.59

2,100.60

12.03

G

Maruti Suzuki India Ltd

1,227.84

8,810.40

7.18

G

Kotak Mahindra Bank Ltd

211.66

1,304.70

6.16

G

HDFC Bank Ltd

423.71

2,063.60

4.87

G

HDFC Ltd

399.58

1,823.55

4.57

G

Motherson Sumi Systems Ltd

74.85

307.65

4.11

G

Tata Consultancy Services Ltd

446.80

1,721.60

3.85

G

 

Given here are the G-Score Calculations for the 10 companies. 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

Tata Consultancy Services Ltd

1

1

1

1

1

1

1

0

7

7.65%

0.5355

HDFC Bank Ltd

1

1

0

1

1

0

1

1

6

7.78%

0.4668

Kotak Mahindra Bank Ltd

1

1

0

1

1

0

1

1

6

7.62%

0.4572

Pidilite Industries Ltd

1

1

0

1

1

1

0

1

6

6.41%

0.3846

HDFC Ltd

1

1

1

1

1

0

1

0

6

6.05%

0.3630

Maruti Suzuki India Ltd

1

1

0

1

1

1

1

1

7

4.51%

0.3157

Bajaj Finance Ltd

1

1

1

1

1

0

0

1

6

5.22%

0.3132

Avenue Supermarts Ltd

1

1

1

1

1

0

0

0

5

4.33%

0.2165

Gruh Finance Ltd

1

1

0

1

1

0

0

0

4

5.03%

0.2012

Motherson Sumi Systems Ltd

1

1

0

1

1

1

1

0

6

3.34%

0.2004

NET G-SCORE

3.4541

There are no Value Stocks and thus we do not calculate the F-Score. 

So, the total score of this Fund is 3.4541.

6.2.1.3.3       ADITYA BIRLA SL FRONTLINE EQUITY FUND

The allocation of funds is as follows:

Equity

96.98%

Rs. 19718.41 crores

Debt

2.34%

Rs. 476.69 crores

Others

0.67%

Rs. 136.50 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

HDFC Bank Ltd

8.67%

1,762.54

ICICI Bank Ltd

5.09%

1,034.93

Infosys Ltd

4.84%

984.55

ITC Ltd

4.57%

929.22

Larsen & Toubro Ltd

2.80%

569.54

Maruti Suzuki India Ltd

2.76%

560.99

Yes Bank Ltd

2.38%

484.97

Mahindra & Mahindra Ltd

2.32%

470.84

HCL Technologies Ltd

2.15%

437.60

HDFC Ltd

2.12%

431.87

 

We separate the Growth Stocks from the Value Stocks.

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

Maruti Suzuki India Ltd

1,227.84

8,810.40

7.18

G

ITC Ltd

38.45

264.55

6.87

G

HDFC Bank Ltd

423.71

2,063.60

4.87

G

HDFC Ltd

399.58

1,823.55

4.57

G

Infosys Ltd

298.73

1,267.40

4.24

G

HCL Technologies Ltd

232.15

933.20

4.03

G

Yes Bank Ltd

111.82

331.65

2.97

G

Larsen & Toubro Ltd

576.44

1,323.20

2.30

V

ICICI Bank Ltd

187.98

293.00

1.56

V

Mahindra & Mahindra Ltd

634.07

913.90

1.44

V

 

Given here are the G-Score Calculations for the 7 companies. 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

HDFC Bank Ltd

1

1

0

1

1

0

1

1

6

8.67%

0.5202

ITC Ltd

1

1

1

1

1

1

1

0

7

4.57%

0.3199

Infosys Ltd

1

0

0

1

0

1

1

0

4

4.84%

0.1936

Maruti Suzuki India Ltd

1

1

0

1

1

1

1

1

7

2.76%

0.1932

HDFC Ltd

1

1

1

1

1

0

1

0

6

2.12%

0.1272

Yes Bank Ltd

1

0

0

1

1

0

0

1

4

2.38%

0.0952

HCL Technologies Ltd

1

0

0

1

1

0

0

0

3

2.15%

0.0645

NET G-SCORE

72.92%

1.5138

Given here are the F-Score Calculations for the 3 companies. 

Company

F1

F2

F3

F4

F5

F6

F7

F8

F9

F-Score

Weight

Contribution

ICICI Bank Ltd

1

1

1

1

1

0

1

0

1

7

5.09%

0.3563

Larsen & Toubro Ltd

1

1

0

1

1

0

0

1

0

5

2.80%

0.1400

Mahindra & Mahindra Ltd

1

1

0

1

1

1

1

0

0

6

2.32%

0.1392

NET F-SCORE

27.08%

0.6355

 So, the total score of this Fund is ((1.5138/7*10) * 72.92% + (0.6355/3*10) * 27.08%) = 2.1506.

6.2.1.3.4       FRANKLIN INDIA BLUECHIP FUND

The allocation of funds is as follows:

Equity

96.68%

Rs. 7830.06 crores

Debt

0.00%

Rs. 0.00 crores

Others

3.32%

Rs. 268.87 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

HDFC Bank Ltd

10.30%

834.39

Infosys Ltd

6.08%

492.72

Larsen & Toubro Ltd

4.87%

394.68

Bharti Airtel Ltd

4.52%

366.13

Yes Bank Ltd

4.27%

346.20

Mahindra & Mahindra Ltd

4.22%

341.49

ICICI Bank Ltd

3.71%

300.09

Axis Bank Ltd

3.37%

272.95

Kotak Mahindra Bank Ltd

3.30%

266.92

State Bank of India Ltd

2.60%

210.25

 

We separate the Growth Stocks from the Value Stocks.

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

Kotak Mahindra Bank Ltd

211.66

1,304.70

6.16

G

HDFC Bank Ltd

423.71

2,063.60

4.87

G

Infosys Ltd

298.73

1,267.40

4.24

G

Yes Bank Ltd

111.82

331.65

2.97

G

Larsen & Toubro Ltd

576.44

1,323.20

2.30

V

Axis Bank Ltd

250.44

522.55

2.08

V

Bharti Airtel Ltd

185.95

370.90

2.00

V

ICICI Bank Ltd

187.98

293.00

1.56

V

Mahindra & Mahindra Ltd

634.07

913.90

1.44

V

State Bank of India Ltd

263.25

276.85

1.05

V

 

Given here are the G-Score Calculations for the 4 companies. 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

HDFC Bank Ltd

1

1

0

1

1

0

1

1

6

10.30%

0.6180

Infosys Ltd

1

0

0

1

0

1

1

0

4

6.08%

0.2432

Kotak Mahindra Bank Ltd

1

1

0

1

1

0

1

1

6

3.30%

0.1980

Yes Bank Ltd

1

0

0

1

1

0

0

1

4

4.27%

0.1708

NET G-SCORE

50.70%

1.2300

Given here are the F-Score Calculations for the 6 companies. 

Company

F1

F2

F3

F4

F5

F6

F7

F8

F9

F-Score

Weight

Contribution

ICICI Bank Ltd

1

1

1

1

1

0

1

0

1

7

3.71%

0.2597

Mahindra & Mahindra Ltd

1

1

0

1

1

1

1

0

0

6

4.22%

0.2532

Larsen & Toubro Ltd

1

1

0

1

1

0

0

1

0

5

4.87%

0.2435

Bharti Airtel Ltd

1

0

1

0

1

0

1

0

1

5

4.52%

0.2260

Axis Bank Ltd

1

0

0

0

1

0

0

1

0

3

3.37%

0.1011

State Bank of India Ltd

1

0

0

0

1

0

1

0

0

3

2.60%

0.0780

NET F-SCORE

49.30%

1.1615

 So, the total score of this Fund is ((1.23/4*10) * 50.7% + (1.1615/6*10) * 49.3%) = 2.5134.

6.2.1.3.5       IDFC CORE EQUITY FUND

The allocation of funds is as follows:

Equity

94.01%

Rs. 2611.71 crores

Debt

7.56%

Rs. 210.13 crores

Others

-1.58%

– Rs. 43.81 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

HDFC Bank Ltd

6.91%

192.00

Infosys Ltd

4.17%

115.82

Larsen & Toubro Ltd

2.84%

78.90

Kotak Mahindra Bank Ltd

2.43%

67.57

ITC Ltd

2.14%

59.35

HDFC Ltd

2.12%

58.84

MRF Ltd

2.05%

56.84

Reliance Industries Ltd

2.01%

55.87

Mahindra & Mahindra Ltd

2.00%

55.50

Jindal Steel & Power Ltd

1.98%

55.06

 

We separate the Growth Stocks from the Value Stocks.

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

ITC Ltd

38.45

264.55

6.87

G

Kotak Mahindra Bank Ltd

211.66

1,304.70

6.16

G

HDFC Bank Ltd

423.71

2,063.60

4.87

G

HDFC Ltd

399.58

1,823.55

4.57

G

Infosys Ltd

298.73

1,267.40

4.24

G

MRF Ltd

20,373.94

74,359.80

3.65

G

Larsen & Toubro Ltd

576.44

1,323.20

2.30

V

Reliance Industries Ltd

501.59

1,015.55

2.02

V

Mahindra & Mahindra Ltd

634.07

913.90

1.44

V

Jindal Steel & Power Ltd

335.48

229.55

0.69

V

 

Given here are the G-Score Calculations for the 6 companies. 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

HDFC Bank Ltd

1

1

0

1

1

0

1

1

6

6.91%

0.4146

Infosys Ltd

1

0

0

1

0

1

1

0

4

4.17%

0.1668

Kotak Mahindra Bank Ltd

1

1

0

1

1

0

1

1

6

2.43%

0.1458

ITC Ltd

1

1

1

1

1

1

1

0

7

2.14%

0.1498

HDFC Ltd

1

1

1

1

1

0

1

0

6

2.12%

0.1272

MRF Ltd

1

1

1

1

1

1

1

1

8

2.05%

0.1640

NET G-SCORE

69.18%

1.1682

Given here are the F-Score Calculations for the 4 companies. 

Company

F1

F2

F3

F4

F5

F6

F7

F8

F9

F-Score

Weight

Contribution

Reliance Industries Ltd

1

1

1

1

1

1

1

1

1

9

2.01%

0.1809

Larsen & Toubro Ltd

1

1

0

1

1

0

0

1

0

5

2.84%

0.1420

Mahindra & Mahindra Ltd

1

1

0

1

1

1

1

0

0

6

2.00%

0.1200

Jindal Steel & Power Ltd

1

0

0

0

1

0

0

1

0

3

1.98%

0.0594

NET F-SCORE

30.82%

0.5023

 So, the total score of this Fund is ((1.1682/6*10) * 69.18% + (0.5023/4*10) * 30.82%) = 1.7340.

6.2.1.3.6       RELIANCE SMALL CAP FUND

The allocation of funds is as follows:

Equity

93.66%

Rs. 6503.81 crores

Debt

6.29%

Rs. 436.79 crores

Others

0.05%

Rs. 3.47 crores

 

So, definitely, this is an Equity Based Fund.

The portfolio of this fund is as follows.

Company

%age Allocation

Value (in Rs. Crores)

HDFC Ltd

2.72%

188.92

VIP Industries Ltd

2.27%

157.61

Navin Fluorine International Ltd

2.26%

156.76

Deepak Nitrite Ltd

2.17%

150.65

Cyient Ltd

2.14%

148.41

Zydus Wellness Ltd

2.04%

141.65

LG Balakrishna & Bros Ltd

1.97%

136.61

RBL Bank Ltd

1.81%

125.91

Magma Fincorp Ltd

1.76%

122.15

Tejas Networks Ltd

1.70%

118.22

 

We separate the Growth Stocks from the Value Stocks.

Company

Book Value / Share

Price Per Share

Price-to-Book Value

G/V

Tejas Networks Ltd

6.76

312.15

46.57

G

VIP Industries Ltd

28.90

433.00

15.02

G

Zydus Wellness Ltd

145.31

1,447.10

9.99

G

RBL Bank Ltd

115.57

542.00

4.72

G

HDFC Ltd

399.58

1,823.55

4.57

G

Deepak Nitrite Ltd

54.86

244.55

4.47

G

Cyient Ltd

188.09

716.25

3.83

G

LG Balakrishna & Bros Ltd

306.67

612.05

2.01

V

Magma Fincorp Ltd

91.67

164.40

1.80

V

Navin Fluorine International Ltd

780.81

678.35

0.87

V

 

Given here are the G-Score Calculations for the 7 companies. 

Company

G1

G2

G3

G4

G5

G6

G7

G8

G-Score

Weight

Contribution

HDFC Ltd

1

1

1

1

1

0

1

0

6

2.72%

0.1632

VIP Industries Ltd

0

1

0

1

1

0

1

1

5

2.27%

0.1135

RBL Bank Ltd

1

1

0

1

1

0

0

1

5

1.81%

0.0905

Deepak Nitrite Ltd

1

0

0

1

1

1

0

0

4

2.17%

0.0868

Cyient Ltd

0

1

0

1

1

1

0

0

4

2.14%

0.0856

Tejas Networks Ltd

1

1

1

1

0

0

1

0

5

1.70%

0.0850

Zydus Wellness Ltd

1

1

0

1

1

0

0

0

4

2.04%

0.0816

NET G-SCORE

71.26%

0.7062

Given here are the F-Score Calculations for the 3 companies. 

Company

F1

F2

F3

F4

F5

F6

F7

F8

F9

F-Score

Weight

Contribution

Navin Fluorine Intl Ltd

1

1

0

1

1

1

1

0

0

6

2.26%

0.1356

LG Balakrishna & Bros Ltd

0

1

0

1

1

0

0

1

1

5

1.97%

0.0985

Magma Fincorp Ltd

1

1

0

1

1

0

0

1

0

5

1.76%

0.0880

NET F-SCORE

28.74%

0.3221

 So, the total score of this Fund is ((0.7062/7*10) * 71.26% + (0.3221/3*10) * 28.74%) = 1.0275.

6.2.1.3.7       FINAL SELECTION OF EQUITY BASED MUTUAL FUNDS

Based on our analysis, we find that the scores of the 6 Mutual Funds are as follows.

FUND

SCORE

Axis Long Term Equity Fund

3.4541

Axis Focused 25 Fund

3.4252

Franklin India Bluechip Fund

2.5134

Aditya Birla SL Frontline Equity Fund

2.1506

IDFC Core Equity Fund

1.7340

Reliance Small Cap Fund

1.0275

In our list and according to our analysis, the top 2 funds are from Axis Mutual Funds. So, we will select 1 of them for diversification between Fund Managers. So, we select Axis Long Term Equity Fund, Franklin India Bluechip Fund and Aditya Birla SL Frontline Equity Fund.

6.2.2  ALLOCATING FUNDS TO MUTUAL FUNDS

We allocate funds to the selected Mutual Funds such that the Treynor Ratio of the portfolio of Mutual Fund is maximised. So, initially allocate equal amounts to each Mutual Fund and then solve the Linear Programming Problem.

The initial set up is as follows.

image009

The LPP Setup is as follows.

image010

The Final allocation of Funds is as follows.

image011

So, our allocation of funds in Mutual Funds is as follows.

MUTUAL FUND

Allocation

HDFC Index Fund NIFTY Plan

Rs. 5,00,000

SBI Dynamic Bond Fund

Rs. 1,60,000

Axis Long Term Equity Fund

Rs. 5,00,000

Franklin India Bluechip Fund

Rs. 3,40,000

Aditya Birla SL Frontline Equity Fund

Rs. 5,00,000

 

6.3    SHARES

We first need to select the Shares to invest in. Once we have selected the Shares to invest in, we need to allocate funds to the same.

6.3.1  SELECTING SHARES

We need to select Shares based on 3 different strategies. We make the selection through of the strategies below.

6.3.1.1              SELECTING SHARES BASED ON SUE SCORE

SUE is the acronym for Standardised Unexpected Earning.

SUE Score = (Actual EPS – Expected EPS) / (Standard Deviation of EPS)

·      EPS stands for Earning Per Share.

·      Actual EPS is the current EPS as released in the last year’s Financial Statements.

·      Expected EPS is calculated by finding the Average of the EPS at the end of the last 4 years (before the current year).

·      Standard Deviation of EPS is calculated by finding the Standard Deviation of the last 4 year’s EPS.

After finding the SUE Score of the Stocks of interest, order them in the descending order of the SUE Score.

As a thumb rule, we must sell all Stocks with negative SUE Score.

Also, we must consider selling all Stocks with very low SUE Score.

We must consider buying the Stock with the highest SUE Scores (Which are positive).

This strategy must be applied immediately after the annual results of the companies are announced. So, this is the ideal time to apply this strategy.

In our case, we are building a new portfolio. So, we will only take the case where we will buy the Shares with high SUE Score.

I have collected the EPS for different companies and arranged them by SUE Score. The data can be seen in the table below.

Stock

 2014 

 2015 

 2016 

 2017 

 Expected EPS 

Standard Deviation

Actual

EPS

 SUE 

DLF Ltd

3.63

3.03

1.72

4.01

3.10

1.0030

25.02

21.8560

Titan Company Ltd

8.28

9.19

7.60

8.01

8.27

0.6740

12.73

6.6168

HDFC Ltd

50.93

55.65

64.50

69.56

60.16

8.4208

96.77

4.3476

Hindustan Unilever Ltd

18.24

20.17

19.22

20.79

19.61

1.1158

24.09

4.0196

Chambal Fertilisers and Chemicals Ltd

5.73

5.90

7.12

5.31

6.02

0.7773

8.70

3.4543

ITC Ltd

7.45

8.04

7.74

8.47

7.93

0.4359

9.24

3.0166

JSW Steel Ltd

1.75

7.32

(1.40)

14.66

5.58

7.0442

25.71

2.8573

Kotak Mahindra Bank Ltd

16.00

19.72

18.86

26.84

20.36

4.6065

32.54

2.6452

Bharat Electronics Ltd

3.45

3.60

4.53

5.07

4.16

0.7709

6.20

2.6432

Infosys Ltd

46.57

54.07

58.96

62.73

55.58

6.9763

73.66

2.5913

Tata Elxsi Ltd

12.06

16.52

24.86

27.83

20.32

7.2956

38.54

2.4977

HDFC Bank Ltd

36.45

42.64

50.63

59.63

47.34

10.0424

71.28

2.3841

Capital First Ltd

7.19

12.56

18.21

24.52

15.62

7.4464

33.09

2.3461

Yes Bank Ltd

6.94

9.56

12.03

14.63

10.79

3.2974

18.38

2.3018

Reliance Industries Ltd

38.25

40.04

50.45

50.53

44.82

6.5908

56.94

1.8393

Jain Irrigation Systems Ltd

(0.86)

1.20

1.02

3.53

1.22

1.7984

4.42

1.7780

Minda Corporation Ltd

3.92

4.32

5.16

4.62

4.51

0.5224

5.37

1.6558

LIC Housing Finance Ltd

26.12

27.65

33.05

38.49

31.33

5.6247

39.91

1.5259

Just Dial Ltd

17.19

19.70

20.55

17.45

18.72

1.6596

21.25

1.5229

HCL Technologies Ltd

46.50

52.04

39.72

60.32

49.65

8.7195

62.63

1.4892

Maruti Suzuki India Ltd

94.47

126.07

181.99

248.67

162.80

67.7249

260.86

1.4479

MindTree Ltd

27.03

32.04

32.93

24.92

29.23

3.8731

34.78

1.4330

Dabur India Ltd

5.24

6.07

7.11

7.25

6.42

0.9451

7.69

1.3464

Tata Consultancy Services Ltd

97.69

101.35

123.20

133.45

113.92

17.2139

135.21

1.2366

Vedanta Ltd

21.24

(52.77)

(41.38)

14.83

(14.52)

37.9681

27.82

1.1151

Hindustan Petroleum Corp Ltd

7.09

9.83

30.68

54.05

25.41

21.8051

47.37

1.0070

Castrol India Ltd

4.80

6.22

6.78

6.99

6.20

0.9867

7.02

0.8336

Suzlon Energy Ltd

(14.16)

(24.70)

1.16

1.71

(9.00)

12.7938

1.26

0.8018

Kiri Industries Ltd

5.22

79.88

73.71

95.66

63.62

40.0138

93.46

0.7458

Bharat Forge Ltd

6.54

11.28

16.38

14.57

12.19

4.3194

15.13

0.6801

ACC Ltd

61.88

31.30

35.05

49.23

44.37

13.9989

51.32

0.4968

India Cements Ltd

(7.92)

0.00

3.80

4.92

0.20

5.8084

2.15

0.3357

Cummins India Ltd

16.70

27.85

26.02

26.56

24.28

5.1130

25.68

0.2733

Jet Airways Ltd

(363.54)

(184.63)

106.66

38.60

(100.73)

214.8889

(56.03)

0.2080

Federal Bank Ltd

4.97

6.17

2.83

5.03

4.75

1.3940

4.74

(0.0072)

Granules India Ltd

3.71

4.45

5.68

7.19

5.26

1.5231

5.22

(0.0246)

Bank of Baroda Ltd

22.81

23.29

17.69

(21.99)

10.45

21.7747

7.88

(0.1180)

Canara Bank Ltd

65.20

55.46

58.64

(46.70)

33.15

53.3876

22.74

(0.1950)

Siemens Ltd

16.94

32.99

81.85

31.93

40.93

28.2490

33.61

(0.2590)

Tata Motors Ltd

43.01

42.99

34.10

21.95

35.51

9.9676

31.13

(0.4397)

REC Ltd

24.01

27.06

28.82

31.97

27.97

3.3283

25.98

(0.5964)

Oil India Ltd

24.87

21.70

17.30

13.28

19.29

5.0670

16.04

(0.6409)

Sintex Industries Ltd

11.52

12.24

13.83

2.46

10.01

5.1267

2.39

(1.4868)

Biocon Ltd

6.90

8.29

9.17

10.20

8.64

1.3982

6.21

(1.7380)

Coal India Ltd

23.92

21.73

22.59

14.93

20.79

4.0108

13.60

(1.7933)

State Bank of India Ltd

18.99

22.76

15.75

0.30

14.45

9.8587

(5.11)

(1.9840)

Karnataka Bank Ltd

13.45

19.51

17.95

16.00

16.73

2.6146

11.52

(1.9917)

Dr. Reddy Laboratories Ltd

115.35

137.11

124.89

77.93

113.82

25.5304

57.06

(2.2232)

Bharti Airtel Ltd

7.55

11.56

15.20

9.51

10.96

3.2695

2.75

(2.5096)

Axis Bank Ltd

26.86

31.42

35.04

16.51

27.46

8.0291

1.78

(3.1980)

Apollo Tyres Ltd

19.94

19.21

22.06

21.59

20.70

1.3465

12.65

(5.9783)

So, we select the following Shares based on SUE Score.

1.   DLF Ltd

2.   Titan Company Ltd

3.   HDFC Ltd

4.   Hindustan Unilever Ltd

6.3.1.2              SELECTING SHARES BASED ON SLOAN SCORE

Based on recommendation from Investment Bankers, I have collected names of some Stocks to invest in. I investigate these Stocks based on their Sloan Score. These Stocks are as follows.

1.   Tata Consultancy Services Ltd (TCS)

2.   Reliance Industries Ltd

3.   Siemens Ltd

4.   Infosys Ltd

5.   ITC Ltd

6.   Karnataka Bank Ltd

7.   LIC Housing Finance Ltd

8.   National Aluminium Company Ltd (NALCO)

6.3.1.2.1       TATA CONSULTANCY SERVICES LTD (TCS)

From the Annual Report of TCS, we gather the following.

Account Head

(All figures are in crores of Rs.)

2016-17

2015-16

Cash and bank balances

1,316

4,806

TOTAL CURRENT ASSETS

68,619

53,377

TOTAL ASSETS

89,758

77,417

Current Liabilities

10,701

11,309

TOTAL SHORT TERM DEBTS

200

113

Tax expense

6,413

6,264

Revenue from Operations

117,966

108,646

Depreciation and amortisation expense

1,575

1,459

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 68,619 – 53,377 = 15,242

DCash = 1,316 – 4,806 = -3,490

DCurrent Liabilities = 10,701 – 11,309 = -608

DTotal Short Term Debts = 200 – 113 = 87

DTax Payable = 6,413 – 6,264 = 149

Accruals (A) 

(15,242 – (-3,490)) – ((-608) – 87 – 149) – 1,575 

18,001

Total Revenue from Continuous Operations (I) 

 

117,966

A Normalised = A / (Average Assets of Last 2 Year)

18,001 / 83,587.50 

0.215

I Normalised = I / (Average Assets of Last 2 Year)

117,966 / 83,587.50 

1.411

Sloan Score 

1.411 – 0.215  

1.196

 

6.3.1.2.2       RELIANCE INDUSTRIES LTD (RIL)

From the Annual Report of RIL, we gather the following.

Account Head

(All figures are in crores of Rs.)

2016-17

2015-16

Cash and bank balances

323.58

241.26

TOTAL CURRENT ASSETS

4,388.24

11,558.64

TOTAL ASSETS

35,480.63

35,363.82

Current Liabilities

2,060.22

2,907.62

TOTAL SHORT TERM DEBTS

0

0

Tax expense

661.15

817.52

Revenue from Operations

9,320.86

8,969.33

Depreciation and amortisation expense

1,443.25

1,232.45

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 4,388.24 – 11,558.64 = -7,170.40

DCash = 323.58 – 241.26 = 82.32

DCurrent Liabilities = 2,060.22 – 2,907.62 = -847.40

DTotal Short Term Debts = 0 – 0 = 0

DTax Payable = 661.15 – 817.52 = -156.37

Accruals (A) 

((-7,170.40) – 82.32) – ((-847.40) – 0 – (-156.37)) – 1,443.25 

-8,004.94

Total Revenue from Continuous Operations (I) 

 

117,966

A Normalised = A / (Average Assets of Last 2 Year)

(-8,004.94) / 35,422.225 

-0.226

I Normalised = I / (Average Assets of Last 2 Year)

9,320.86 / 35,422.225

0.263

Sloan Score 

0.263 – (-0.226)  

0.489

 

6.3.1.2.3       SIEMENS LTD

From the Annual Report of Siemens, we gather the following.

Account Head

(All figures are in lakhs of Rs.)

2016-17

2015-16

Cash and bank balances

8,375

10,604

TOTAL CURRENT ASSETS

58,429

55,329

TOTAL ASSETS

1,33,804

1,25,717

Current Liabilities

43,394

42,916

TOTAL SHORT TERM DEBTS

5,447

6,206

Tax expense

2,042

1,773

Revenue from Operations

83,049

79,644

Depreciation and amortisation expense

3,211

2,764

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 58,429 – 55,329 = 3,100

DCash = 8,375 – 10,604 = -2,229

DCurrent Liabilities = 43,394 – 42,916 = 478

DTotal Short Term Debts = 5,447 – 6,206 = -759

DTax Payable = 2,042 – 1,773 = 269

Accruals (A) 

(3,100 – (-2,229)) – (478 – (-759) – 269) – 3,211 

1,150

Total Revenue from Continuous Operations (I) 

 

83,049

A Normalised = A / (Average Assets of Last 2 Year)

1,150 / 1,29,760.5 

0.009

I Normalised = I / (Average Assets of Last 2 Year)

83,049 / 1,29,760.5

0.640

Sloan Score 

0.640 – 0.009  

0.631

 

6.3.1.2.4       INFOSYS LTD

From the Annual Report of Infosys, we gather the following.

Account Head

(All figures are in crores of Rs.)

2016-17

2015-16

Cash and bank balances

16,770

19,153

TOTAL CURRENT ASSETS

44,090

47,682

TOTAL ASSETS

75,877

79,885

Current Liabilities

11,662

11,786

TOTAL SHORT TERM DEBTS

0

0

Tax expense

4,003

5,068

Revenue from Operations

61,941

59,289

Depreciation and amortisation expense

1,863

1,703

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 44,090 – 47,682 = -3,592

DCash = 16,770 – 19,153 = -2,383

DCurrent Liabilities = 11,662 – 11,786 = -124

DTotal Short Term Debts = 0 – 0 = 0

DTax Payable = 4,003 – 5,068 = -1,065

Accruals (A) 

((-3,592) – (-2,383)) – ((-124) – 0 – (-1,065)) – 1,863 

-4,013

Total Revenue from Continuous Operations (I) 

 

61,941

A Normalised = A / (Average Assets of Last 2 Year)

-4,013 / 77,881

-0.052

I Normalised = I / (Average Assets of Last 2 Year)

61,941 / 77,881

0.795

Sloan Score 

0.795 – (-0.052)

0.847

 

6.3.1.2.5       ITC LTD

From the Annual Report of ITC, we gather the following.

Account Head

(All figures are in crores of Rs.)

2016-17

2015-16

Cash and bank balances

156.15

75.79

TOTAL CURRENT ASSETS

26,269.10

24,862.50

TOTAL ASSETS

55,943.27

51,691.88

Current Liabilities

9,238.39

8,499.89

TOTAL SHORT TERM DEBTS

0.01

3.60

Tax expense

6,830.07

6,354.27

Revenue from Operations

55,448.46

51,944.57

Depreciation and amortisation expense

1,152.79

1,077.40

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 26,269.10 – 24,862.50 = 1,460.60

DCash = 156.15 – 75.79 = 80.36

DCurrent Liabilities = 9,238.39 – 8,499.89 = 738.50

DTotal Short Term Debts = 0.01 – 3.60 = -3.59

DTax Payable = 6,830.07 – 6,354.27 = 475.80

Accruals (A) 

(1,460.60 – 80.36) – (738.50 – (-3.59) – 475.80) – 1,152.79 

-92.84

Total Revenue from Continuous Operations (I) 

 

55,448.46

A Normalised = A / (Average Assets of Last 2 Year)

-92.84 / 53,817.575

-0.002

I Normalised = I / (Average Assets of Last 2 Year)

55,448.46 / 53,817.575

1.030

Sloan Score 

1.030 – (-0.002)

1.032

 

6.3.1.2.6       KARNATAKA BANK LTD

From the Annual Report of Kar Bank, we gather the following.

Account Head

(All figures are in lakhs of Rs.)

2016-17

2015-16

Cash and bank balances

2,645.62

2,488.45

TOTAL CURRENT ASSETS

36,947.37

34,294.15

TOTAL ASSETS

56,500.33

51,836.60

Current Liabilities

2,321.53

2,438.93

TOTAL SHORT TERM DEBTS

1,051.48

1,037.76

Tax expense

185.23

330.97

Revenue from Operations

4,992.21

4,698.42

Depreciation and amortisation expense

242.14

203.33

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 36,947.37 – 34,294.15 = 2,653.22

DCash = 2,645.62 – 2,488.45 = 157.17

DCurrent Liabilities = 2,321.53 – 2,438.93 = -117.40

DTotal Short Term Debts = 1,051.48 – 1,037.76 = 13.72

DTax Payable = 185.23 – 330.97 = -145.74

Accruals (A) 

(2,653.22 – 157.17) – ((-117.40) – 13.72 – (-145.74)) – 242.14 

2,239.29

Total Revenue from Continuous Operations (I) 

 

4,992.21

A Normalised = A / (Average Assets of Last 2 Year)

2,239.29 / 54,168.465

0.041

I Normalised = I / (Average Assets of Last 2 Year)

4,992.21 / 54,168.465

0.092

Sloan Score 

0.092 – 0.041

0.051

 

6.3.1.2.7       LIC HOUSING FINANCE LTD

From the Annual Report of LIC HFL, we gather the following.

Account Head

(All figures are in lakhs of Rs.)

2016-17

2015-16

Cash and bank balances

2,529.23

3,284.44

TOTAL CURRENT ASSETS

2,958.10

3,762.75

TOTAL ASSETS

4,574.13

4,857.26

Current Liabilities

848.26

1,201.66

TOTAL SHORT TERM DEBTS

216.86

249.11

Tax expense

-79.96

13.00

Revenue from Operations

333.69

0.09

Depreciation and amortisation expense

11.02

12.19

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 2,958.10 – 3,762.75 = -804.65

DCash = 2,529.23 – 3,284.44 = -755.21

DCurrent Liabilities = 848.26 – 1,201.66 = -353.40

DTotal Short Term Debts = 216.86 – 249.11 = -32.25

DTax Payable = -79.96 – 13.00 = -92.96

Accruals (A) 

((-804.65) – (-755.21)) – ((-353.40) – (-32.25) – (-92.96)) – 11.02 

167.73

Total Revenue from Continuous Operations (I) 

 

333.69

A Normalised = A / (Average Assets of Last 2 Year)

167.73 / 4,715.695

0.036

I Normalised = I / (Average Assets of Last 2 Year)

333.69 / 4,715.695

0.071

Sloan Score 

0.071 – 0.036

0.035

 

6.3.1.2.8       NATIONAL ALUMINIUM COMPANY LTD

From the Annual Report of NALCO, we gather the following.

Account Head

(All figures are in crores of Rs.)

2016-17

2015-16

Cash and bank balances

2,287.23

5,103.15

TOTAL CURRENT ASSETS

5,655.79

7,343.65

TOTAL ASSETS

14,501.65

16,710.19

Current Liabilities

2,651.93

1,981.95

TOTAL SHORT TERM DEBTS

0

0

Tax expense

219.52

366.93

Revenue from Operations

8,050.02

7,269.23

Depreciation and amortisation expense

480.36

426.12

 

 

Based on these figures, we can calculate the components of the Sloan Score.

DCurrent Assets = 5,544.79– 7,343.65 = -1,687.86

DCash = 2,287.23 – 5,103.15 = -2,815.92

DCurrent Liabilities = 2,651.93 – 1,981.95 = 669.98

DTotal Short Term Debts = 0 – 0 = 0

DTax Payable = 219.52 – 366.93 = -147.41

Accruals (A) 

((-1,687.86) – (-2,815.92)) – (669.98 – 0 – (-147.41)) – 11.02 

-169.69

Total Revenue from Continuous Operations (I) 

 

8,050.02

A Normalised = A / (Average Assets of Last 2 Year)

-169.69 / 15,605.92

-0.011

I Normalised = I / (Average Assets of Last 2 Year)

8,050.02 / 15,605.92

0.516

Sloan Score 

0.516 – (-0.011)

0.527

 

6.3.1.2.9       MAKING THE SELECTION

We arrange the companies by the Sloan Score in the descending order.

Company

Sloan Score

Percentile

Tata Consultancy Services Ltd

1.196

100

ITC Ltd

1.032

88

Infosys Ltd

0.847

75

Siemens Ltd

0.631

63

National Aluminium Company Ltd

0.527

50

Reliance Industries Ltd

0.489

38

Karnataka Bank Ltd

0.051

25

LIC Housing Finance Ltd

0.035

13

So, we select the following Shares based on SLOAN Score.

1.   Tata Consultancy Services Ltd

2.   ITC Ltd

6.3.1.3              SELECTING SHARES BASED ON PAIR TRADING

With the Government working on providing Electricity to All, the Power Generating Companies are expected to do well in the coming years. So, for Pair Trading, we will try to select a Pair from the Power Generating Companies.

An allied industry to the Power Generation Industry is the Coal Industry. 37% of the fuel requirement of the Power Generating Companies is met by the Coal industry. So, when the Coal becomes costlier, Power Industry has adverse effect and the vice versa.

So, we will form a Pair between Power Generating Companies and the Coal Producing Companies.

6.3.1.3.1       FORMING PAIR FROM POWER GENERATING INDUSTRY

The Stocks being studied are as follows:

1.   Gujarat Industries Power Company Ltd. (GIPCL)

2.   Jaiprakash Power Ventures Ltd. (JPVL)

3.   JSW Energy Ltd. (JSW)

4.   NHPC Ltd. (NHPC)

5.   NTPC Ltd. (NTPC)

6.   Power Grid Corporation India Ltd. (PGRID)

7.   Reliance Industries Ltd. (RIL)

8.   SJVN Ltd. (SJVN)

 

The following steps were followed:

1.   The Historical Prices from 1-Jun-2017 to 31-May-2018 were collected for each of the Stocks.

2.   From the Historical Prices gathered in Step 1, the normalised prices of each day for each Stocks was calculated as (Stock Price – Mean(Stock Prices)) / Standard Deviation(Stock Prices).

3.   From the normalised prices determined in Step 2, the distance was calculated for each day for each pair by squaring the difference between the normalised prices for the day of each pair.

4.   Lastly, the square of the distances determined in Step 3 were summed to find the distance between each pair. This figure is provided in the table below.

 

GIPCL

JPVL

JSW

NHPC

NTPC

PGRID

RIL

SJVN

GIPCL

 

 

 

 

 

 

 

 

JPVL

73.93

 

 

 

 

 

 

 

JSW

373.48

252.88

 

 

 

 

 

 

NHPC

379.05

450.83

607.67

 

 

 

 

 

NTPC

270.75

230.36

199.22

620.60

 

 

 

 

PGRID

367.39

495.96

780.77

402.08

474.66

 

 

 

RIL

484.97

599.09

861.60

210.93

686.01

193.01

 

 

SJVN

356.87

253.64

145.23

493.60

338.50

746.79

722.27

 

 

The least distance is found between GIPCL and JPVL.

So, we form a pair between Gujarat Industries Power Company Ltd and Jaiprakash Power Ventures Ltd.

6.3.1.3.2       FORMING PAIR BETWEEN POWER GENERATING INDUSTRY AND COAL INDUSTRY

The Stocks from Power Industry being studied are as follows:

1.   Gujarat Industries Power Company Ltd. (GIPCL)

2.   Jaiprakash Power Ventures Ltd. (JPVL)

3.   JSW Energy Ltd. (JSW)

4.   NHPC Ltd. (NHPC)

5.   NTPC Ltd. (NTPC)

6.   Power Grid Corporation India Ltd. (PGRID)

7.   Reliance Industries Ltd. (RIL)

8.   SJVN Ltd. (SJVN)

The Stocks from Coal Industry being studied are as follows:

1.   Coal India Ltd (CIL)

2.   Gujarat Mineral Development Corporation Ltd (GMDC)

3.   Gujarat Natural Resources Ltd (GNRL)

4.   Auroma Coke Ltd (ACL)

5.   Austral Coke & Projects Ltd (ACPL)

The following steps were followed:

1.   The Historical Prices from 1-Jun-2017 to 31-May-2018 were collected for each of the Stocks.

2.   From the Historical Prices gathered in Step 1, the normalised prices of each day for each Stocks was calculated as (Stock Price – Mean(Stock Prices)) / Standard Deviation(Stock Prices).

3.   From the normalised prices determined in Step 2, the distance was calculated for each day for each pair by squaring the difference between the normalised prices for the day of each pair.

4.   Lastly, the square of the distances determined in Step 3 were summed to find the distance between each pair. This figure is provided in the table below.

 

CIL

GMDC

GNRL

ACL

ACPL

GIPCL

552.48

123.18

345.02

288.26

100.93

JPVL

435.14

145.78

425.48

181.98

150.66

JSW

142.16

270.44

680.09

118.87

516.34

NHPC

686.09

328.67

180.44

712.26

242.99

NTPC

412.42

208.79

545.87

148.91

438.62

PGRID

841.18

477.22

278.55

656.26

399.97

RIL

830.04

559.33

201.94

866.16

347.25

SJVN

150.68

248.28

557.56

261.02

408.01

 

The minimum distance is between GIPCL and ACPL. However, ACPL is not a solvent company. Thus, we will not form this pair.

The second minimum distance is between JSW and ACL. However, ACL is not a solvent company. Thus, we will not form this pair.

The third minimum distance is between GIPCL and GMDC. However, GIPCL already appears in our pair from Power Industry. Thus, we will not form this pair.

The fourth minimum distance is between JSW Energy Ltd and Coal India Ltd. We will form this pair.

 

6.3.2  ALLOCATING FUNDS TO EACH SHARE

We apply Capital Asset Pricing Model (CAPM). According to CAPM, the risk of investment is measured in b. We gather the b for all the Shares we have selected and then apply Linear Programming to maximise b within the limits as acceptable to the Investor.

A value of 1 for b indicates that the Shares moves along with the market. So, the risk in the Share is that same as the risk in the market.

A value higher than 1 for b indicates that the Share is expected to move higher than the market. This indicates that the Share is riskier than the market. Investing in this Share, we should thus expect higher returns as we are taking additional risk.

If b is less than 0, it indicates that the Share moves in opposite direction of the market.

After discussion with the Investor, it was established that they are ready for a maximum value of 1.6 for b. The values of b is listed below.

Share

b

DLF Ltd

2.24

Titan Company Ltd

1.10

HDFC Ltd

0.86

Hindustan Unilever Ltd

0.65

Tata Consultancy Services Ltd

0.25

ITC Ltd

0.78

JSW Energy Ltd

1.74

Coal India Ltd

0.71

Gujarat Industries Power Company Ltd

1.49

Jaiprakash Power Ventures Ltd

2.29

 

Initially, we make equal allocation of funds to each of the Shares.

image012

The Linear Programming Problem is set up as follows.

image013

The Final Allocation is as follows.

image014

So, we decide the allocation in Shares as follows.

Share

Funds

Current Market Price

Possible Number of Shares

DLF Ltd (DLF)

Rs. 4,00,000

198.35

2,017

Titan Company Ltd (Titan)

Rs. 2,25,400

900.00

250

HDFC Ltd (HDFC)

Rs. 2,14,000

1826.50

117

Hindustan Unilever Ltd (HUL)

Rs. 93,000

1623.50

57

Tata Consultancy Services Ltd (TCS)

Rs. 1,10,400

1841.20

60

ITC Ltd (ITC)

Rs. 82,600

264.85

312

JSW Energy Ltd (JSWE)

Rs. 3,60,200

70.65

5,098

Coal India Ltd (CIL)

Rs. 71,600

275.30

260

Gujarat Industries Power Company Ltd (GIPCL)

Rs. 2,82,800

92.00

3,074

Jaiprakash Power Ventures Ltd (JPPV)

Rs. 1,60,000

3.25

49,231

Note that JSW Energy Ltd and Coal India Ltd are pairs. So, we must consider the sum of their allocation for trading in this pair.

The same applies for Gujarat Industries Power Company Ltd and Jaiprakash Power Ventures Ltd.

We study the performance possible from this portfolio of Shares. From www.economictimes.com, I have gathered the historical returns from the selected shares. The calculations from these figures is provided below. (For calculating Sharpe Ratio and Treynor Ratio, we consider Risk Free Rate as 7.10%)

Share

RETURNS (%)

Er (%)

s 

(%)

Sharpe

Ratio

Treynor

Ratio

1 yr

3 yr

5 yr

DLF

3.98

86.77

5.80

32.18

47.28

0.53

0.11

Titan

73.12

150.03

300.45

174.53

115.63

1.45

1.52

HDFC

11.41

50.84

118.82

60.36

54.33

0.98

0.62

HUL

45.88

86.80

170.30

100.99

63.41

1.48

1.44

TCS

51.86

44.80

152.79

83.15

60.41

1.26

3.04

ITC

-12.50

31.53

20.01

13.01

22.83

0.26

0.08

JSWE

12.96

-26.97

47.64

11.21

37.34

0.11

0.02

CIL

7.94

-30.39

-9.40

-10.62

19.19

-0.92

-0.25

GIPCL

-12.05

13.73

31.92

11.20

22.09

0.19

0.03

JPPV

-24.18

-49.53

-85.45

-53.05

30.79

-1.95

-0.26

EXPECTED PORTFOLIO RETURNS

41.37

 

 

 

PORTFOLIO RISK (s)

28.60

 

 

PORTFOLIO SHARPE RATIO

1.20

 

PORTFOLIO TREYNOR RATIO

0.23

The diversification can be studied from the correlations between the Shares. They are as follows.

 

DLF

Titan

HDFC

HUL

TCS

ITC

JSWE

CIL

GIPCL

JPPV

DLF

 

 

 

 

 

 

 

 

 

 

Titan

-0.16

 

 

 

 

 

 

 

 

 

HDFC

-0.13

1.00

 

 

 

 

 

 

 

 

HUL

-0.17

1.00

1.00

 

 

 

 

 

 

 

TCS

-0.53

0.92

0.91

0.93

 

 

 

 

 

 

ITC

0.72

0.57

0.60

0.56

0.21

 

 

 

 

 

JSWE

-0.88

0.62

0.59

0.63

0.87

-0.29

 

 

 

 

CIL

-0.90

-0.28

-0.31

-0.27

0.11

-0.95

0.58

 

 

 

GIPCL

0.12

0.96

0.97

0.96

0.78

0.78

0.37

-0.54

 

 

JPPV

0.08

-1.00

-1.00

-1.00

-0.64

-0.64

-0.55

0.36

-0.98

 

 

7       FINAL RECOMMENDATION

The final recommendation from this analysis is that the company should invest its Rs. 50,00,000 as follows to get an expected return of at least 18% per annum in 5 years.

Instrument

Amount (Rs.)

Fixed Deposits

Rs. 10,00,000

HDFC Bank Ltd

Rs. 2,60,000

Ananda Cooperative Bank

Rs. 1,10,000

Gnana Shale Souharda Cooperative Bank

Rs. 6,30,000

Mutual Funds

Rs. 20,00,000

HDFC Index Fund NIFTY Plan

Rs. 5,00,000

SBI Dynamic Bond Fund

Rs. 1,60,000

Axis Long Term Equity Fund

Rs. 5,00,000

Franklin India Bluechip Fund

Rs. 3,40,000

Aditya Birla SL Frontline Equity Fund

Rs. 5,00,000

Shares

Rs. 20,00,000

DLF Ltd

Rs. 4,00,000

Titan Company Ltd

Rs. 2,25,400

HDFC Ltd

Rs. 2,14,000

Hindustan Unilever Ltd

Rs. 93,000

Tata Consultancy Services Ltd

Rs. 1,10,400

ITC Ltd

Rs. 82,600

JSW Energy Ltd

Rs. 3,60,200

Coal India Ltd

Rs. 71,600

Gujarat Industries Power Company Ltd

Rs. 2,82,800

Jaiprakash Power Ventures Ltd

Rs. 1,60,000

The Investor needs staying invested in the Fixed Deposits and Mutual Funds for 5 years or more.

In case of the Shares, the Investor need staying invested in the following shares for 1 year before changing the portfolio- DLF Ltd, Titan Company Ltd, HDFC Ltd, Hindustan Unilever Ltd, Tata Consultancy Services Ltd, ITC Ltd. The remaining 4 shares combination needs changing after 6 months.

8       APPENDIX I – Annual Reports Studied

The Annual Reports of the following Companies were studied in preparing this report.

Sl.

Company

1.     

ACC Ltd

2.     

Apollo Tyres Ltd

3.     

Auroma Coke Ltd

4.     

Austral Coke & Projects Ltd

5.     

Avenue Supermarts Ltd

6.     

Axis Bank Ltd

7.     

Bajaj Finance Ltd

8.     

Bajaj Finserv Ltd

9.     

Bank of Baroda Ltd

10.  

Bharat Electronics Ltd

11.  

Bharat Forge Ltd

12.  

Bharti Airtel Ltd

13.  

Biocon Ltd

14.  

Canara Bank Ltd

15.  

Capital First Ltd

16.  

Castrol India Ltd

17.  

Chambal Fertilisers and Chemicals Ltd

18.  

Coal India Ltd

19.  

Cummins India Ltd

20.  

Cyient Ltd

21.  

Dabur India Ltd

22.  

Deepak Nitrate Ltd

23.  

DLF Ltd

24.  

Dr. Reddy Laboratories Ltd

25.  

Federal Bank Ltd

26.  

Granules India Ltd

27.  

Gruh Finance Ltd

28.  

Gujarat Industries Power Company Ltd

29.  

Gujarat Mineral Development Corporation Ltd

30.  

Gujarat Natural Resources Ltd

31.  

HCL Technologies Ltd

32.  

HDFC Bank Ltd

33.  

HDFC Ltd

34.  

Hindustan Petroleum Corp Ltd

35.  

Hindustan Unilever Ltd

36.  

ICICI Bank Ltd

37.  

India Cements Ltd

38.  

Infosys Ltd

39.  

ITC Ltd

40.  

Jain Irrigation Systems Ltd

41.  

Jaiprakash Power Ventures Ltd

42.  

Jet Airways Ltd

43.  

Jindal Steel & Power Ltd

44.  

JSW Energy Ltd

45.  

JSW Steel Ltd

46.  

Just Dial Ltd

47.  

Karnataka Bank Ltd

48.  

Kiri Industries Ltd

49.  

Kotak Mahindra Bank Ltd

50.  

Larsen & Toubro Ltd

51.  

LG Balakrishna & Bros Ltd

52.  

LIC Housing Finance Ltd

53.  

Magma Fincorp Ltd

54.  

Mahindra & Mahindra Ltd

55.  

Maruti Suzuki India Ltd

56.  

Minda Corporation Ltd

57.  

MindTree Ltd

58.  

Motherson Sumi Systems Ltd

59.  

National Aluminium Company Ltd

60.  

Navin Fluorine International Ltd

61.  

NHPC Ltd

62.  

NTPC Ltd

63.  

Oil India Ltd

64.  

Pidilite Industries Ltd

65.  

Power Grid Corporation India Ltd

66.  

RBL Bank Ltd

67.  

REC Ltd

68.  

Reliance Industries Ltd

69.  

Shree Cement Ltd

70.  

Siemens Ltd

71.  

Sintex Industries Ltd

72.  

SJVN Ltd

73.  

State Bank of India Ltd

74.  

Supreme Industries Ltd

75.  

Suzlon Energy Ltd

76.  

Tata Consultancy Services Ltd

77.  

Tata Elxsi Ltd

78.  

Tata Motors Ltd

79.  

Tejas Networks Ltd

80.  

Titan Company Ltd

81.  

Vedanta Ltd

82.  

VIP Industries Ltd

83.  

Yes Bank Ltd

84.  

Zydus Wellness Ltd

9       APPENDIX II – List of Web Sites Searched

Sl.

Paper

1.    

www.economictimes.com

2.    

www.moneycontrol.com

3.    

www.bseindia.com

4.    

www.nseindia.com

5.    

www.wikipedia.com

 

10   APPENDIX III – List of Papers Researched

Sl.

Paper

1.    

http://blog.efpsa.org/2013/02/28/how-to-read-and-get-the-most-out-of-a-journal-article/

 

This paper explains how to read papers on trading strategies.

2.    

https://www.chicagobooth.edu/~/media/FE874EE65F624AAEBD0166B1974FD74D.pdf

 

This paper is the original paper by Piotroski (F-Score).

3.    

https://www.chicagobooth.edu/pdf/ballbrown1968.pdf

 

This paper is the original paper by Ball & Brown.

4.    

http://econ.au.dk/fileadmin/Economics_Business/Education/Summer_University_2012/6308_Advanced_Financial_Accounting/Advanced_Financial_Accounting/2/Sloan_1996_TAR.pdf

 

This paper is the original paper by Sloan.

5.    

http://pages.stern.nyu.edu/~lpederse/papers/BettingAgainstBeta.pdf

 

This paper is the original paper by Pedersen and Frazzini.

6.    

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=299107

 

This paper is the original paper by Jagdeesh and Titman.

7.    

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=403180

 

This paper is the original paper by Professor Mohanram (G-Score).

8.    

https://www.fidelity.com/viewpoints/guide-to-diversification

 

This paper explains the importance of Diversification.

9.    

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=141615

 

This paper by Gatev, Goetzmann and Rouwenhorst forms the basis for this “Pairs Trading” strategy.

10.                  

https://blog.wealthfront.com/benchmark-investments-portfolio-performance/

 

This paper discusses the ways in which one can benchmark one’s portfolio performance.

 

11   BIBLIOGRAPHY

TERM

EXPLANATION

CAPM

The capital asset pricing model (CAPM) is a model used to determine a theoretically appropriate required rate of return of an asset, to make decisions about adding assets to a well-diversified portfolio.

The model takes into account the asset’s sensitivity to non-diversifiable risk (also known as systematic risk or market risk), often represented by the quantity beta (β) in the financial industry, as well as the expected return of the market and the expected return of a theoretical risk-free asset. CAPM assumes a particular form of utility functions (in which only first and second moments matter, that is risk is measured by variance, for example a quadratic utility) or alternatively asset returns whose probability distributions are completely described by the first two moments (for example, the normal distribution) and zero transaction costs (necessary for diversification to get rid of all idiosyncratic risk). Under these conditions, CAPM shows that the cost of equity capital is determined only by beta. Despite it failing numerous empirical tests, and the existence of more modern approaches to asset pricing and portfolio selection (such as arbitrage pricing theory and Merton’s portfolio problem), the CAPM still remains popular due to its simplicity and utility in a variety of situations.

F-Score

Piotroski F-Score is a number between 0-9 which is used to assess strength of company’s financial position. The Score is used by financial investors in order to find the best value stocks (nine being the best). The Score is named after Stanford Accounting Professor, Joseph Piotroski.

Fixed Deposit

fixed deposit (FD) is a financial instrument provided by banks or NBFCs (Non-Banking Financial Company) which provides investors a higher rate of interest than a regular savings account, until the given maturity date. It may or may not require the creation of a separate account. It is known as a term deposit or time deposit in Canada, Australia, New Zealand, and the US, and as a bond in the United Kingdom and India. They are considered to be very safe investments. Term deposits in India, Nepal, and Pakistan are used to denote a larger class of investments with varying levels of liquidity. The defining criteria for a fixed deposit is that the money cannot be withdrawn from the FD as compared to a recurring deposit or a demand deposit before maturity. Some banks may offer additional services to FD holders such as loans against FD certificates at competitive interest rates. It’s important to note that banks may offer lesser interest rates under uncertain economic conditions. The interest rate varies between 4 and 7.25 percent. The tenure of an FD can vary from 7, 15 or 45 days to 1.5 years and can be as high as 10 years. These investments are safer than Post Office Schemes as they are covered by the Deposit Insurance and Credit Guarantee Corporation (DICGC). However, DICGC guarantees amount up to  100,000 (about $1555) per depositor per bank. They also offer income tax and wealth tax benefits.

Mutual Fund

mutual fund is a professionally managed investment fund that pools money from many investors to purchase securities. These investors may be retail or institutional in nature.

Mutual funds have advantages and disadvantages compared to direct investing in individual securities. The primary advantages of mutual funds are that they provide economies of scale, a higher level of diversification, they provide liquidity, and they are managed by professional investors. On the negative side, investors in a mutual fund must pay various fees and expenses.

Share

Corporations issue shares which are offered for sale to raise share capital. The owner of shares in the corporation is a shareholder (or stockholder) of the corporation. A share is an indivisible unit of capital, expressing the ownership relationship between the company and the shareholder. The denominated value of a share is its face value, and the total of the face value of issued shares represent the capital of a company, which may not reflect the market value of those shares.

The income received from the ownership of shares is a dividend. The process of purchasing and selling shares often involves going through a stockbroker as a middle man. There are different types of shares such as equity shares, preference shares, bonus shares, right shares, employees stock option plans and sweat equity shares.

Sharpe Ratio

The Sharpe Ratio (also known as the Sharpe Index, the Sharpe Measure, and the reward-to-variability ratio) is a way to examine the performance of an investment by adjusting for its risk. The ratio measures the excess return (or risk premium) per unit of deviation in an investment asset or a trading strategy, typically referred to as risk, named after William F. Sharpe.

Treynor Ratio

The Treynor Ratio (sometimes called the reward-to-volatility ratio or Treynor Measure), named after Jack L. Treynor, is a measurement of the returns earned in excess of that which could have been earned on an investment that has no diversifiable risk (e.g., Treasury bills or a completely diversified portfolio), per each unit of market risk assumed.

The Treynor ratio relates excess return over the risk-free rate to the additional risk taken; however, systematic risk is used instead of total risk. The higher the Treynor ratio, the better the performance of the portfolio under analysis.

 


Advertisements

This site uses Akismet to reduce spam. Learn how your comment data is processed.