Sie sind auf Seite 1von 13

ASSIGNMENT 3

APPLIED ECONOMETRICS

SUBMITTED TO : PROF PC PADHAN


SUBMITTED BY: TEJASWINI KONDA :H16056
Dataset

Dependent Variable: Aurobindo Pharma Ltd companys monthly share price for the past 5
years was taken as the Dependent Variable.
Independent Variable:
1. Volume of the stock traded: Volume of the stock traded per month over the last 5
years was taken as one of the independent variables.
2. Dollar Rate: The fluctuations in the US dollar value on a monthly basis over the last
5 years was taken.
3. Nifty Index: The monthly Nifty data over the last 5 years.
4. Inflation:

Theoretical Relationship:

1.Volume - it tells whether there are buyers or sellers for this stock in the market and Price
tells the direction in which the stock is moving. The relation between volume and price of a
stock is a very important aspect of technical analysis as it is used to confirm the trends and
chart patterns. A higher volume determines stronger movement of of the stock influencing its
price.
Volume is an important aspect of technical analysis because it is used to confirm trends and
chart patterns. Any price movement up or down with relatively high volume is seen as a
stronger, more relevant move than a similar move with weak volume.

Therefore, if you are looking at a large price movement, you should also examine the volume
to see whether it tells the same story.

There are a several combinations possible

Increase in price and Increasing volume: The increase in price is directly


proportional to the increase in the volume as it means that a larger group of people are
investing in the stock. However, there is another possibility
Declining Price and Declining volume: A declining price means that there are lesser
number of sellers as compared to the previous price. The price of the stock might not
increase but it just stagnates there.
Increasing price and Declining volume: There are lesser number of buyers in the
market so the price would be high.
Steady Price and Increasing Volume: This indicates a distribution, that means
resistance. It signifies that a big seller is likely in the market. There is no concrete way
to tell if the buyers able to drive the price higher, or if they will sell the stock and the
stock eventually reverses.
Increasing Price with a Steady Volume: This indicates that there might be a who is
steadily buying shares without trying to attract much attention.

2.Dollar Rate

Changes in exchange rates directly influence the competitiveness of a firm because any
fluctuations in exchange rate affect the value of the earnings and cost of its funds as
Aurobindo Pharma being a leader in generic drug manufacturing with 70% of its products
exported. Exchange value plays a critical role in determining the Stock price. For critical raw
material, the company sources it from other countries. For this foreign currencies are required
to fund the operations.
A depreciation of the rupee makes a better price for its goods and leads to an increase in
foreign demand. An appreciation of the rupee decreases profits for this firm as it
predominantly exports its products it leads to a decrease in foreign demand of its products.
On a macro basis, the impact of exchange rate fluctuations on stock prices depend on 1. The
contribution of a countrys international trades in its economy 2. The degree of the trade
imbalance. Portfolio balance stresses the role of capital account transaction. A blooming stock
market attracts capital flows from foreign investors, which causes an increase in the demand
for Rupee. Reverse in case of a falling stock price.

3.Nifty Index-

Aurobindo Pharma is in the list of 50 companies that comprise NIFTY. Its weightage it 0.49.
The Nifty index gives the picture of the economic conditions. Any fluctuations in the top 50
companies would affect the overall economy of the country as these companies are the top
contributors in their respective sectors So fluctuations in the nifty value are directly
correlated with the stock price.

4.Inflation-
Rising inflation could sometimes have a positive effect of consumers investing more money
in stocks than holding it in the form as cash. However, it is not a simple examination of the
investing power of the consumer but a dynamic economic scenario that has a multiplier
effect. Inflation would increase the production, procurement costs, and directly hit the units
sold by the manufacturer. So its impact on the stock price has a varying impact. This
parameter was chosen to study this interesting relationship.

1.Multiple Linear Regression:

Multiple linear regression model was used for predictive analysis. It was used to explain the
relationship between the continuous dependent variable and the the continuous independent
variables. Also to determine the future trends. The significance of the effects of the
independent variable on the dependent variable was tested with the help of the value of the p
values for each of the coefficients. The hypothesis formulated is as follows

A ) T- test:

Hypothesis

Null Hypothesis: The coefficient of the independent variable is 0

Alternative Hypothesis: The coefficient of the independent variable is not zero.

; with a 5% significance level

So, if the p-value < 0.05 the hypothesis is rejected and the coefficient of the independent
variable is significant.
Test Results 1

Inference

The p value of the inflation is >0.05 it means the null hypothesis that the coefficient value is
0 is accepted. It means that the effect of inflation on the stock price estimation can be
ignored.

Action

The regression was re-run removing infaltion.


Test Results 2

Inference

All the null hypothesis are rejected and all the coefficients are significant so the model is

Price= -2077.312 + 30.89309 *Dollar_Rate+ 0.093865 * Nifty_Price -1.17E-06

B )F test

F-test was done to check the significance of R-Square. The null hypothesis being R- square is
zero , alternate hypothesis is R Square value is not zero. From the table the p value of the
test <5% therefore the null hypothesis is rejected, hence R square value is not zero. From
the table Adjusted R Square is 0.83883 which means that the predicted model explains
83.8837% of the relationship, which is very significant.
2.MULTI-COLLINEARITY:

CORRELATION MATRIX

LINEAR REGRESSION

Inference 1: It can be inferred that the independent variable inflation is highly


correlated with the independent variable nifty price. From the linear regression
results we see that the p-value of the independent variable is 0.1611 which is
much greater than 0.05 which means that the null hypothesis of the coefficient
being insignificant is accepted. So the variable is dropped.

After Removing the Variable Inflation: Correlation matrix


After dropping the variable inflation It can be seen that the multicollinearity
problem is eliminated.

3. HETEROSKEDASTICITY:

In statistics the collection of random variables is heteroscedasticity. If the sub populations


have different variances, i. e the variance is not constant in the data the problem of
heteroscedasticity is observed. Heteroscedasticity means that there is no constant variance. It
is a concern in regression analysis as it can invalidate the model predicted.in
homoscedasticity it is assumed that the errors are not correlated and uniform. But in
heteroscedastity they are not. They vary with time and fit into an equation. The
OlS method of estimation fails and Gauss Markov Theorem cannot be applied, the estimators
do not follow the best linear unbiased estimators and their variances are very high, it does not
affect the OLS co-efficient but effects the OLS variance of the coefficients can be biased. In
the case of heteroscedasticity as the actual variance and covariance are underestimated. The
predictive power of the regression model reduces.
To correct this asymptotic values of the errors are substituted and then checked for
heteroscedasticity

Hypothesis

Null hypothesis: Homoscedasticity


Alternative hypothesis: Heteroscedasticity
To test the Heteroscedasticity two tests were performed
White Test
Breusch-Pagan_Godfrey Test
Inference:
From the test results it can be seen that the p value obtained using the BPG test is
0.0039 which is less than 0.05, which means the null hypothesis of
Homoscedasticity is no longer valid.
Which means that the predicted model might not be valid for future.

Corrective Method:

To eliminate the problem White method is used to obtain the asymptotic


standard errors and Coefficients during regression .
4. AUTO-CORRELATION:

There are two types of auto-correlation one is the inter -correlation with the time series data.
It is the degree of similarity between the a time series and the lagged version of it. So it is n-
1th time period influencing the nth time period. Post detection to treat this the variable
showing auto- correlation, a new variable with Ynew= Y-1 is included and multiple
regression is run again.
Second is the correlation with the other independent variables.

Null Hypothesis: No auto correlation


Alternative hypothesis : Auto correlation present

Test:

From the test results we can see that the p value of the test is less than 5% it means
the null hypothesis is rejected, so there is Auto correlation

Remedy:
To treat the auto-correlation the lag values should be taken as a separate variable.

Test Results: The following are the test results post the corrective action of
including the lag variable as Lag_Price = Price (-1)

Inference: It can be seen post the remedial action the observed R square value is
0.7584 means that the null hypothesis of the autocorrelation being absent is
accepted. Thus, the autocorrelation problem has been treated.

4.REDUNDANT VARIABLE TEST

Redundant variable test is performed to check if any variable in the


model is redundant and needs tobe excluded. Here the Inflation is
taken as a redundant variable.

Null hypothesis: The variable is redundant


Alternative Hypothesis : The variable is not redundant
5.OMITTED VARIABLE

It is used to check it the omitted variable from the model is redundant or non redundant.

Null Hypothesis: The omitted is not a omitted variable


Alternative Hypothesis: The omitted variable is a omitted variable

Omitted variable for testing is INFLATION


The p value is >5% it means the null hypothesis that it is not an omitted
variable is accepted. So inflation can be omitted