Project: Return on Indian Stocks in COVID-19 times

Problem Statement: Collect the daily price information of the following five Indian stocks for the period 1
February-31 July, 2020 and convert them to daily returns. The stocks to be studied are Reliance Industries
Limited (RELIANCE), Tata Steel Limited (TATASTEEL), Adani Enterprises Limited (ADANIENT), Aditya Birla
Capital (ABCAPITAL) and Infosys Limited (INFY).

Solutions:

(a) What is the probability distribution of the daily returns of the five stocks over this six-month period?

In normal distribution most of the values lie around the mean and by plotting the daily return of each
against their probability, we can assume that the returns are normally distributed with different mean and
standard deviation (Please refer the tables shown below). The graphs of their PDF and CDF are
approximately normally distributed (graphs follow in part (b)).

ADANIENT ABCAPITAL
X P(X) CDF Frequency
X P(X) CDF Frequency
0.0
<-0.17 0.01 1 1 <-0.132 0.008 0.008 1
0.0 -0.10-0.07 0.025 0.033 3
-0.13-0.11 0.01 2 1
-0.07-0.04 0.083 0.117 10
0.0
-0.11-0.09 0.01 3 1 -0.04-0.02 0.183 0.300 22
0.0 -0.02-.008 0.317 0.617 38
-0.09-0.07 0.03 6 4 0.008-
0.0 0.03 0.242 0.858 29
-0.07-0.05 0.02 8 2 0.036-
0.1 0.06 0.100 0.958 12
-0.05-0.03 0.07 4 8 0.064-
0.3 0.09 0.025 0.983 3
-0.03-0.01 0.17 1 20 0.092-
0.6 0.12 0.017 1.000 2
-0.01-0.009 0.31 2 37
0.8
0.009-0.03 0.24 6 29
0.9
0.03-0.05 0.09 5 11
0.9
0.05-0.07 0.03 8 4
1.0
0.07-0.09 0.02 0 2

INFY RELIANCE
X P(X) CDF Frequency X P(X) CDF Frequency
<-0.1003 0.01 0.01 1 <-0.13 0.01 0.01 1
-0.1003-0.0773 0.02 0.03 2 -0.13--0.1 0.01 0.02 1
-0.0773-0.0543 0.03 0.05 3 -0.1--0.07 0.02 0.03 2
-0.0543-0.0313 0.05 0.10 6 -0.07--
-0.0313-0.0082 0.21 0.31 25 0.04 0.03 0.07 4
-0.0083-0.0147 0.43 0.74 52 -0.04--
0.0147-0.0377 0.17 0.91 20 0.01 0.22 0.28 26
0.0377-0.0607 0.03 0.94 4 -0.01-0.02 0.47 0.75 56
0.0607-0.0837 0.03 0.98 4 0.02-0.05 0.19 0.94 23
0.0837-0.1067 0.02 0.99 2 0.05-0.08 0.02 0.96 2
0.1067-0.1297 0.01 1.00 1 0.08-0.11 0.03 0.98 3
0.11-0.14 0.01 0.99 1
0.14-0.17 0.01 1.00 1
 TATASTEEL
X P(X) CDF Frequency
<-0.11 0.01 0.01 1
-0.11--
0.08 0.03 0.04 4
-0.08--
0.05 0.05 0.09 6
-0.05--
0.02 0.13 0.23 16
-0.02-0.01 0.43 0.65 51
0.01-0.04 0.25 0.90 30
0.04-0.07 0.08 0.98 9
0.07-0.1 0.02 0.99 2
0.13-0.16 0.01 1.00 1
(b). Are the probability distributions of the daily returns of the five
stocks similar?

Though all the stock returns are assumed to be normally distributed about their mean value, the stock
returns of ADANIENT appear skewed to the left (negative skewness). The remaining stocks (i.e. ABCAPITAL,
INFY, TATASTEEL, RELIANCE) do not exhibit skew. When the PDF and CDF of stock returns are plotted for
individual stocks, the distribution is reminiscent of a normal distribution, as seen below.

INFY - Probability Distribution of Daily ABCAPITAL - Probability Distribution of

Returns Daily Returns
1.00 1.000
0.900
0.80 0.800
0.60 0.700
0.600
0.40 0.500
0.20 0.400
0.300
0.00 0.200
03 773 543 313 082 147 377 607 837 067 067 0.100
.10 .0 .0 .0 .0 .0 .0 .0 .0 .1 .1 0.000
0 --0 3--0 3--0 3--0 83-0 47-0 77-0 07-0 37-0 >0
<- 3 <- - - - -0.02- 0.008- 0.036- 0.064- >0.092
00 077 054 031 .00 01 03 06 08 0.132 0.104- 0.076- 0.048- 0.008 0.036 0.064 0.092
-0
.1
-0
.
-0
.
-0
. -0 0. 0. 0. 0. -0.076 -0.048 -0.02

P(X) CDF P(X) CDF

ADANIENT - Probibility Distribution of TATASTEEL- Probability Distribution of

Daily Returns Daily Returns
1.00
0.90 1.00
0.80 0.80
0.70
0.60 0.60
0.50
0.40 0.40
0.30
0.20 0.20
0.10
0.00 0.00
<-0.11 -0.11-- -0.08-- -0.05-- -0.02- 0.01- 0.04- 0.07- >0.13
17 .1
1
.0
9
.0
7
.0
5
.0
3
.0
1 09 .03 .05 .07 .07 0.08 0.05 0.02 0.01 0.04 0.07 0.1
- 0. --0 --0 --0 --0 --0 --0 -0.0 9-0 3-0 5-0 >0
< 13 11 .09 .07 .05 .03 .01 .00 0.0 0.0
. .
-0 -0 -0 -0 -0 -0 -0 0

P(X) CDF P(X) CDF

 RELIANCE - Probability Distribution of Daily Returns
1.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
<-0.13 -0.13--0.1 -0.1--0.07 -0.07--0.04 -0.04--0.01 -0.01-0.02 0.02-0.05 0.05-0.08 0.08-0.11 0.11-0.14 >0.14

P(X) CDF

Another method of deducing similarity is by looking at the correlation between each pair of stock returns.
(shown below)

Cor(INFY, ABCAPITAL) 0.354 Cor(ABCAPITAL, TATASTEEL) 0.518

Cor(INFY, ADANI) 0.591 Cor(ABCAPITAL, RELIANCE) 0.365
Cor(INFY, TATASTEEL) 0.527 Cor(ADANI, TATASTEEL) 0.518
Cor(INFY, RELIANCE) 0.528 Cor(ADANI, RELIANCE) 0.542
Cor(ABCAPITAL, ADANI) 0.538 Cor(TATASTEEL, RELIANCE) 0.594

Interpretation of these co-efficient values shows that stock returns are moderately correlated.
(Assumption: Correlation coefficients between 0.9 and 1.0 indicate high correlation, while correlation
coefficients between 0.5 and 0.7 indicate moderate correlation.)

(c). Calculate the expected daily return and the standard deviation of the daily returns of these five
stocks. Are these similar? If not, what similarities/differences do you observe.
The expected daily return and the standard deviation of the daily returns of the five stocks are as follows:

INFY ABCAPITAL ADANIENT TATASTEEL RELIANCE

Expected Daily Return 0.24% -0.34% -0.12% -0.07% 0.42%
Standard Deviation 3.27% 3.88% 3.64% 3.76% 3.77%

Observations:
From the above values of expectation and standard deviation, we observe that in terms of magnitude of
these two parameters, the five stocks are quite similar.
However, the expected return of INFY and RELIANCE are positive, while the remaining are negative,
implying that only INFY and RELIANCE are expected to give positive returns across this entire period.
The standard deviations of these stocks are quite similar. Except for INFY, all other stocks have SD in the
range of 3.64% to 3.88%, while INFY has a slightly lower SD of 3.27%. Lower standard deviation means the
stock is less volatile (least among the five) and slightly better for risk-averse investors.
(d). An investor is keen to know “how much” she can lose due to daily stock movement. To answer this
question, she wants to compute the largest value of t such that P(X<=t) <= 0.01 where X is the daily
return of the stock. Compute the value of t for each of the five stocks.

We have approximated all probability distributions to normal distributions, and we have to find the largest
value of t such that P(X ≤ t) ≤ 0.01. (Assumption: Relative frequencies are equal to the corresponding true
probabilities.)

  RELIANCE TATASTEEL INFY ABCAPITAL ADANIENT

Mean 0.42% -0.07% 0.24% -0.34% -0.12%
Standard Deviation 3.77% 3.76% 3.27% 3.88% 3.64%
Random variable X ~ N(0.42,3.77) X ~ N(-0.07, 3.76) X ~ N(0.24,3.27) X ~ N(-0.34,3.88) X ~ N(-0.12, 3.64)
P(X ≤ t) ≤ 0.01 -8.34% -8.81% -7.37% -9.36% -8.59%

(e). Now obtain the value of ‘t’ using the one-sided Chebyshev’s inequality (a.k.a. Cantelli’s inequality)
for each of these stocks. Comment on your findings.

Using Cantelli’s inequality: P(X ≤ E(X)- (v × SD(X))) ≤ 1/(1+v 2)

1/(1+v2) = 0.01
v = 9.95

For INFY
t = -0.12-9.95*3.64
Mean= 0.24% t (ADANIENT) = -36.338
Standard Deviation= 3.27%
t = 0.24 – 9.95*3.27
t (INFY) = -32.3
For TATASTEEL
Mean= -0.07
For ABCAPITAL Standard Deviation= 3.76%
t = -0.07-9.95*3.76
Mean= -0.34% t (TATASTEEL) = -37.482
Standard Deviation= 3.88%
t = -0.34-9.95*3.88
t (ABCAPITAL) = -38.946
For RELIANCE
Mean= 0.42
For ADANIENT Standard Deviation= 3.77
t = 0.42-9.95*3.77
Mean= -0.12% t (RELIANCE) = -37.092
Standard Deviation= 3.64%

Value of t
Assuming Normal Distribution Using Chebyshev’s (Cantelli’s) Inequality
INFY -7.37% -32.3
 ABCAPITAL -9.36% -38.946
ADANIENT.NS -8.59% -36.338
TATASTEEL.NS -8.81 -37.482
RELIANCE.NS -8.34 -37.092
A side-by-side comparison of the values obtained via the two methods is displayed below:

Observations:
There is a significant difference in the values of t obtained using Cantelli’s inequality and the value of t
obtained
Stock Weight Expected Return Standard Deviation assuming a
Infosys 0.2 0.24% 0.0327
normal
AB Capital 0.2 -0.34% 0.0388
Adani Enterprises 0.2 -0.12% 0.0364
distribution.
Tata Steel 0.2 -0.07% 0.0376 Value of t
Reliance 0.2 0.42% 0.0377
given by
Cantelli’s
inequality for INFY is 4.38 times that computed assuming a normal distribution, 4.16 for ABCAPITAL, 4.25
for TATASTEEL and so on. In general, all values of t for the 5 stocks obtained by Cantelli’s inequality are
around 4 times that computed assuming a normal distribution. This implies that the probability
distributions of stock returns for the 5 stocks are similar. The difference in values is on account of inability
to gauge a precise nature of distribution for the stock returns. This also suggests that extreme values like
-37.092 can occur more often than assumed by normal distribution.
This also shows that Cantelli’s inequality isn’t very accurate for small sample sets. It would be more
accurate for a larger sample.

(f). A portfolio of these stocks is sought to be created with 20% weightage given to each of these stocks.
What would be the expected daily return of this portfolio? What would be the standard deviation of the
daily return?

Given below is the portfolio consisting of 5 stocks with 20% weights given to each of them –

1. The Expected Return of the portfolio

= w1*E1 + w2*E2 + w3*E3 + w4*E4 + w5*E5
= 0.0239%.

2. The Variance of the 5 stock portfolio

= w1²σ1² + w2²σ2² + w3²σ3² + w4²σ4² + w5²σ5² + 2w1w2σ1σ2ρ(1,2) + 2w1w3σ1σ3ρ(1,3) +
2w1w4σ1σ4ρ(1,4) + 2w1w5σ1σ5ρ(1,5) + 2w2w3σ2σ3ρ(2,3) + 2w2w4σ2σ4ρ(2,4) +
2w2w5σ2σ5ρ(2,5) + 2w3w4σ3σ4ρ(3,4) + 2w3w5σ3σ5ρ(3,5) + 2w4w5σ4σ5ρ(4,5)
 = 0.000813.

3. The Standard Deviation of the portfolio

= Variance0.5
= 0.0285

Refer to the attached Excel file (QM1a_D3_Assignment2.xlsx) for detailed calculations.