See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/301826963

Stock Returns and Volatility: Evidence from Nifty 50 Index

Article  in  Asian Journal of Research in Social Sciences and Humanities · January 2016

DOI: 10.5958/2249-7315.2016.00127.1

CITATIONS READS

0 466

2 authors:

Karthika Palanisamy Parthasarathy Karthikeyan

Vidya Vardhaka College Of Engineering Kongu Engineering College
17 PUBLICATIONS   9 CITATIONS    57 PUBLICATIONS   75 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

A Study on Comparative Analysis of Risk and Return with reference to Selected stocks of BSE Sensex index, India. View project

uncertainities of Aviation View project

All content following this page was uploaded by Karthika Palanisamy on 16 July 2020.

The user has requested enhancement of the downloaded file.

 Asian Journal
Asian Research Consortium of Research in
Social Sciences
and
Asian Journal of Research in Social Sciences and Humanities
Vol. 6, No. 5, May 2016, pp. 434-442. Humanities
ISSN 2249-7315 www.aijsh.com
A Journal Indexed in Indian Citation Index
DOI NUMBER: 10.5958/2249-7315.2016.00127.1

Stock Returns and Volatility:

Evidence from Nifty 50 Index

Karthika Palanisamy*; Karthikeyan Parthasarathy**

*Assistant Professor,
Department of Management Studies,
MPNMJ Engineering College,
Chennimalai, Erode, Tamilnadu, India.
**Assistant Professor,
School of Management Studies,
Kongu Engineering College,
Perundurai, Erode, Tamilnadu, India.

Abstract
This paper investigates the volatility of Nifty 50 index returns over the period of January 2015 to
December 2015 through comparison of various conditional - Heteroskedasticity models. The
empirical results shows that there is significant evidence for asymmetry in stock returns, which is
captured by a EGARCH and TGARCH model, while there is strong persistence of shocks into
volatility. The study finds out the GARCH (1, 1) model is the fitted model for capturing the
symmetric volatility and EGARCH (1, 1) model is fitted model for measuring the asymmetric
volatility.

Keywords: Nifty Return, Volatility, GARCH, Leverage Effect, Volatility Clustering.

________________________________________________________________________________

1. Introduction
Volatility means the amount of uncertainty about changes in a security’s value. A higher volatility
means a security’s value can possibly be spread over a larger range of values whereas, lower
volatility means a security’s value should not fluctuate radically, but changes in value over a period
of time. Over the last few years, modelling volatility of a financial time series has become an
important area and has gained a great deal of attention from academics, researchers and others. The

434
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

time series are found to depend on their own past value depending on past information and exhibit
non-constant variance (Karunanithy Banumathy&Ramachandran Azhagaiah2015). Many empirical
studies about the returns in the stock markets have provided indication in favor of a non-constant
variance existence (Fama 1965; Hathaway 1986). Volatility clustering is refers to large changes
tend to be followed by large changes, of either sign, and small changes tend to be followed by small
changes (Mandelbrot, 1963). The research fails to account the risk level changes and its effect on
expected return. Hence the possible measurement of the volatility of the stock return is important
because investor make decision based on their perception about the volatility in future.

The variance and standard deviation are measure of how spread out a distribution is. In
other words they are measure of variability. The Engle introduced the AutoregressiveConditional
Heteroscedasticity model in the year 1982 for modeling the conditional variance. A generalized
Arch (GARCH) model extended by Bollerslev (1986) is another popular model for estimating
stochastic volatility. These models are widely used in various branches of econometrics, especially
in financial time series analysis. Besides, with the introduction of models of ARCH and GARCH,
there have been number of empirical applications of modelling volatility of financial time series.
However, the GARCH cannot account for leverage effect, however they account for volatility
clustering and leptokurtosis in a series, this necessitated to develop new and extended models over.

The GARCH-in-mean (GARCH-M) model adds a heteroscedasticity term into the mean
equation. GARCH-M is a variation under GARCH model is used to identify the risk return
relationship. The Exponential GARCH model is proposed by Nelson in the year 1991, which is the
logarithmic expression of the conditional volatility used to capture the asymmetric effects. In the
year 1994 Zekoian introduced the Threshold GARCH model, which is used to identify the relation
between asymmetric volatility and return. The TGARCH model also called as GJR model. The
various models were designed by many authors to clearly model and forecast the Time-varying
conditional variance of the time series. Hence, the present paper has been aimed at modelling the
volatility of Indian stock market by the use of different GARCH family models and provides
empirical evidence on the fit of conditional volatility model for the Indian stock market.

2. Review of Literature
Trilochan Tripathy & Luis Gil-Alana (2010) has examined the suitability of various volatility
forecasting model for National Stock Exchange of India. The data included in the study were the
daily closing, high, low and open values of the NSE returns from 2005 to 2008. The study consist
five models which were Historical/Rolling Window Moving Average Estimator, Exponentially
Weighted Moving Average (EWMA), GARCH models, Extreme Value Indicators (EVI) and
Volatility Index (VIX). The model comparison was done on the basis of which models were
explained the ex-post volatility well. The study was revealed that the GARCH and VIX models to
be the best methods for volatility forecasting based on Wald’s constant test. The study concluded
that due to low frequency data extreme value models fail to forecast.

Karunanithy Banumathy & Ramachandran Azhagaiah (2015) has analyzed volatility

pattern of Nifty index based on daily closing prices of S&PCNX Nifty index for ten years period
from January 2003 to December 2012. The study used symmetric GACRH and asymmetric
GARCH models. The study concluded the GARCH (1, 1) model is the best model for measuring

435
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

symmetric volatility effect and TGARCH (1, 1) Model is fitted for measuring asymmetric volatility
effects as.

3. Methodology

3.1. Objectives of the Study

The primary objective of the study is to fit appropriate GARCH model to estimate market volatility
based on Nifty 50 index.

3.2. Research Methodology

Methodology is the systematic, theoretical analysis of the methods applied to a field of study. In
this study various statistical tools namely ADF test, ARCH LM test and GARCH models are
applied. The E-views 9 software is used to analyze the data. The nifty data is collected from NSE
website for the period from January 2015 to December 2015. The Nifty return is calculated as a log
of first difference of daily closing price, which is as follows:

𝐩𝐭
𝐫𝐭 = 𝐥𝐨𝐠
𝐩𝐭−𝟏

Where rt is the log retun of Nifty; pt is closing price at time t;

3.3. Tools used for the Study

Symmetric Volatility Measurement

GARCH

The Garch model (Bollerslev 1986), which allows the conditional variance to be dependent upon
previous own lags, conform to the conditional variance equation in the simplest form as

Mean equation: rt = μ + εt

Variance equation: σ2t = ω + αε2t−1 + βσε2t−1

Where ω >0, α1 ≥ 0, β1 ≥ 0, and rtis the return of the asset at time t, μ is the average return, and εt
is the residual return.The size of parameters α and β determine the short-run dynamics ofthe
volatility time series. If the sum of the coefficient is equal to one, thenany shock will lead to a
permanent change in all future values. Hence,shock to the conditional variance is ‘persistence.’

GARCH-M

In GARCH model, the conditional variance enters the mean equation directly, which is known
generally as a GARCH-M model. The return of a security may depend on its volatility and hence a
simple GARCH-M (1, 1) model can be written as:

436
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

Mean equation: rt = μ + +λ σ2t + εt

Variance equation: σ2t = ω + αε2t−1 + βσε2t−1

The parameter λ in the mean equation is called the risk premium. A positive λ indicates that the
return is positively related to its volatility, i. e. a rise in mean return is caused by an increase in
conditional variance as a proxy of increased risk

Asymmetric Volatility Measures

The main drawback of symmetric GARCH is that the conditional variance is unable to respond
asymmetrically to rise and fall in the stock returns. Hence, number of models have been introduced
to deal with the issue and are called asymmetric models viz., EGARCH, TGARCH and PGARCH,
which are used for capturing the asymmetric phenomena. To study the relation between asymmetric
volatility and return, THE EGARCH (1, 1) and TGARCH (1, 1) models are used in the study.

E-GARCH

This model is based on the logarithmic expression of the conditional variability. The presence of
leverage effect can be tested and this model enables to find out the best model, which capture the
symmetries of the Indian stock market (Nelson 1991) and hence the following equation:

𝛆𝐭−𝟏 𝛑 𝛆𝐭−𝟏
𝐥𝐧 𝛔𝟐𝐭 = 𝛚 + 𝛃𝟏 𝐥𝐧 𝛔𝟐𝐭−𝟏 + − −𝛄
𝛔𝐭−𝟏 𝟐 𝛔𝐭−𝟏

The left-hand side is the log of the conditional variance. The coefficient γ is known as the
asymmetry or leverage term. The presence of leverage effects can be tested by the hypothesis thatγ
<0. The impact is symmetric if γ 0.

T-GARCH

The generalized specification of the threshold GARCH for the conditional variance (Zakoian 1994)
is given by:

σ2t = ω + α1 ε2t−1 + γdt−1 ε2t−1 + β1 σ2t−1

The γ is known as the asymmetry or leverage parameter. In thismodel, good news (εt−1 >0) and the
bad news (εt−1 <0) have differential effect on the conditional variance. Good news has an impact
of αi, while bad news has impact on αi+γi. Hence, if γ is significant and positive, negative shocks
have a larger effect on σ2t than the positive shocks

4. Empirical Analysis and Discussion

Descriptive statistics on Nifty return are presented in table 1. The mean return is negative,
indicating the fact that price has decreased over the period. The descriptive statistics shows that the
returns are negatively skewed, indicating that there is a high probability of earning returns which is
higher than the mean return. The Kurtosisof the series is greater than 3, which implies that the

437
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

return series is fat tailed and does not follow a normal distribution and is further confirmed by
Jarque-Bera test statistics, which is significant at 1 level and hence the null hypothesis of normality
is rejected.

Table 1: Descriptive Statistics of Nifty 50 Daily Return

Mean -0.011608 Minimum -5.915096 Kurtosis 7.0313

Median -0.015308 Std. Dev 1.022012 Jarque-Bera 205.84
Maximum 2.616716 Skewness -0.968169 N 247

The stationary property of the data is very important for time series analysis, before
applying any model the data involved examining the stationarity property by applying the unit root
test. This study applied the Augmented Dickey Fuller Test (ADF). The ADF test values is greater
than the critical value at 1%, 5% and 10% level indicated that the data is stationarity at level series
I(0). After the ADF test the residuals are generated by running the ordinary least square regression
analysis. The ARCH LM test is conducted to the residual part for check the heteroscedasticity
issue.

Table 2:Unit Root Test and ARCH-LM Test for Residuals

ADF Test Probability

T-Statistics -15.00657 0.0000
Critical Value
1% -3.456840
5% -2.873093
10% -2.573002
ARCH LM Test Statistics 0.501585 0.0000

The heteroscedasticity means the variance of the residuals are not constant over the period
of time. The table 2 indicated that the error term has the arch effect because the p value is less than
0.05. The generated residuals values from the regression equations are plotted into graph for the
further analysis. The figure 1 shows that the variability clustering of Nifty 50 returns vary over
period of time and appear in clusters i.e. volatility may be high for certain time periods and low for
other periods. Plots of nifty 50 returns show the volatility clustering phenomenon, where large
changes in returns are likely to be followed by further large changes. Moreover, volatility seems to
react differently to a big price increase or a big price drop, referred to as the leverage effect.

438
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

-2
4
-4
2
-6
0

-2

-4

-6
I II III IV
2015

Residual Actual Fitted

Figure 1: Volatility Clustering of Nifty 50 Daily Returns

After volatility clustering is confirmedwithNifty 50 return series and stationarity using
ADF test, heteroscedasticity effect using ARCH-LM test, the study focuses on determining the best
fitted GARCH model to the return series. Therefore, GARCH model is used for modelling the
volatility of return series in the Indian stock market.

The result of GARCH (1, 1) and GARCH-M (1, 1) models is shown in table.3, which
reveals the parameter of GARCH is statistically significant. In other words, the coefficients viz.,
constant (ω), arch term (α), GARCH term (β) are highly significant at 1% level. In the conditional
variance equation, the estimated β coefficient is considerably greater than α efficient which
resembles that the market has a memory longer than one period and that volatility is more sensitive
to its lagged values than it is to new surprises in the market values.

Table 3: Result of GARCH (1, 1) and GARCH-M (1, 1) Model

Coefficient GARCH(1,1) GARCH-M(1, 1)
Mean Equation
Constant (µ) -0.011317 -0.541699
Risk Premium(λ) - 0.536232
Variance
Constant(ω) 0.122520 0.105044
ARCH Effect(α) 0.037931 0.040665
GARCH Effect(β) 0.838434 0.852582
α+β 0.876365 0.893247
Log Likelihood 352.4975 -350.9641
AIC 2.876619 2.882300
SIC 2.943451 2.953340
ARCH-LM test for Heteroscedasticity
ARCH- LM test 0.4487 0.4242
Probability Chi-square (1) 0.4466 0.4221

439
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

It shows that the volatility is persistent. The sizes of the parameters α and β determine the
volatility in time series. The sum of these coefficients (α andβ) is 0.876365, which is close to unity
indicating that the shock will persist too many future periods. Since the risk-return parameter is
positive and significant at 1% level, it shows that there is a positive relationship between risk and
return. Further, ARCH-LM test is employed to check ARCH effect in residuals of estimated model
and from the results, it is inferred that the p >0.05, which led to conclude that the null hypothesis of
‘no arch effect’ is accepted. In other words, the test statistics do not support for any additional arch
effect remaining in the residuals of the models, which implies that the variance equation is well
specified for the market.

The GARCH-M (1, 1) model is estimated by allowing the mean equation of the return
series to depend on a function of the conditional variance. The constant in mean equation is
significant at 5% level, indicating that there is an abnormal return for the market. From the table 3,
it is inferred that the coefficient of conditional variance (λ) in the mean equation value is positive
however, it is statistically insignificant, which implies that there is no significant impact of
volatility on the expected return, indicating lack of risk-return trade off over time. In the variance
equation of GARCH-M (1, 1), the parameters viz.,ω, α and β are highly significant at 1% level. The
sum of α and β is 0.893247, which infers that shocks will persist in the future period. However, the
ARCH-LM test is applied on residuals and shows that the test statistics do not exhibit additional
arch effect for the entire study period indicating that the variance equation is well specified

Table 4: Result of EGARCH (1, 1) and TARCH (1, 1) Model

Coefficient EGARCH (1, 1) TGARCH (1, 1)

Mean Equation
Constant (µ) -0.135217 -0.030218
Variance
Constant 0.089503 0.103691
ARCH Effect(α) -0.105408 -0.106732
GARCH Effect(β) 0.977829 0.898037
Leverage Effect(γ) -0.160580 0.182854
α+β 0.872421 0.791305
Log Likelihood -337.2564 -348.6089
AIC 2.771307 2.863230
SIC 2.842347 2.934270
ARCH-LM test for Heteroscedasticity
ARCH- LM test 0.2442 0.4181
Probability Chi-square (1) 0.2424 0.4160

In order to capture the asymmetries in the return series, two models have been used viz.,
EGARCH (1, 1) and TGARCH (1, 1). γcaptures the asymmetric effect in both EGARCH (1,1) and
TGARCH (1,1) models.

The asymmetrical EGARCH (1,1) model is used to estimate the returns of the Nifty index
and the result is presented in table 4. The table reveals that arch (α) and GARCH coefficient (β) are
values are close to one, reporting that conditional variance is volatile; the estimated coefficients are
statistically significant at 1% level. γ, the leverage coefficient, is negative and is statistically

440
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

significant at 1% level, exhibiting the there is aleverage effect in return during the study period.
The analysis reveals that there is a negative correlation between past return and future return;
hence, EGARCH (1,1) model supports for the presence of leverage effect on the Nifty 50 index
return. Finally, the ARCH-LM test statistics reveals that the null hypothesis of no heteroscedasticity
in the residuals is accepted.

An another model to test for asymmetric volatility in the Nifty return is TGARCH, which
shows the estimated result of TGARCH (1,1) model. In it, the coefficient of leverage effect (γ) is
positive and significant at 1% level, which implies that negative shocks or bad news have a greater
effect on the conditional variance than the positive shocks or good news. The diagnostic test is
performed to test whether the residuals are normally distributed. The ARCH-LM test statistic for
TGARCH (1, 1) model does not show any additional arch effect present in the residuals of the
model,which implies that the variance equation is well specified for the Indian stock market.

5. Conclusion
In this study, volatility of Nifty 50 index return is tested using the symmetric and asymmetric
GARCH models. The daily closing prices of Nifty index for one year are collected and modelled
using four different GARCH models that capture the volatility clustering and leverage effect for the
study period i. e. from 1st January 2015 to 31st December 2015. GARCH (1,1), GARCH-M (1,1),
EGARCH (1,1), AND TGARCH (1,1) models are employed in the study after confirming the unit
root rest, volatility clustering and arch effect. The results show that the coefficient has the expected
bothsign in the EGARCH and in the TGARCH models. Finally, to identify the best fitted model
among the different specifications of GARCH models, Akaike Information Criterion (AIC) and
Schwarz Information Criterion (SIC) are used, which prove that GARCH (1,1)model has been
found to be the best fitted model among all to capture the symmetric effect as per AIC and SIC
criterion. Further, EGARCH (1,1) model is found to be the best fitted model to capture the
asymmetric volatility based on the highest log likelihood ratios and minimum AIC and SIC
criterion.

References
Bollerslev 1986, ‘Generalized Autoregressive Conditional Heteroscedasticity’, Journal of
Econometrics, vol. 31, pp. 307-327.

Karthikeyan, P., & Mohanasundaram, T. (2012). FII flows and Indian equity market
performance. Asian Journal of Managerial Science, 1(1), 12-16.

Karunanithy Banumathy & Ramachandran Azhagaiah 2015, Modelling Stock Market Volatility:
Evidence from India’, Managing Global Transitions, vol. 13, no. 1, pp. 27-42.

Mandelbrot, B. B., The Variation of Certain Speculative Prices, The Journal of Business 36, No. 4,
(1963), 394-419.

441
 Palanisamy & Parthasarathy (2016). Asian Journal of Research in Social Sciences and Humanities,
Vol. 6, No.5, pp. 434-442

Mohanasundaram, T., & Karthikeyan, P. (2012). Impact of Institutional Investments and Macro-
Economic Variables in the Indian Equity Market. International Research Journal of
Finance and Economics, 94, 69-78.

Nelson, DB 1991, ‘Conditional Heteroscedasticity in Asset Returns: A New Approach’

Econometrica, vol. 59, no. 2, pp. 347–370

TrilochanTripathy& Luis A Gil-Alana 2010, ‘Suitability of Volatility Models for Forecasting Stock
Market Returns: A Study on the Indian National Stock Exchange’, American Journal of
Applied Sciences, vol. 7, no. 11, pp. 1487-1494

Zakoian, JM, 1994, ‘Threshold Heteroscedasticity Models’, Journal of Economic Dynamics and
Control , vol. 18, no.5, pp. 931–944

Zhang Zhichao& Pan Hongyu 2006, ‘Forecasting Financial Volatility: Evidence from Chinese
Stock Market’, Durham Business School Working Paper No. 06/02. Available from:
SSRN: http://ssrn.com/abstract=903937(22 May 2015)

Zhao, XQ 1999, ‘Stock prices, Inflation and Output: Evidence from China’, Applied Economic
Letters, vol. 6 no. 8, pp. 509–511.

442

View publication stats