INTRODUCTION:
Market Indices were primarily constructed to
understand the sentiments of market and how various factors are affecting the
market as a whole. The volatility and return depicted in Indices gives a bird
eye view of how the sector or market is performing. In India Sensex has widely
been used as an indicator of Economy.
It broadly reflects the impact of various
macroeconomic variables. S and P Sensex measures
the performance of thirty largest companies having highest market capitalization
and are generally to most financially sound and liquid companies and has
representation across all major sectors of economy it broadly covers 50% of
total market cap of Bombay Stock Exchange and can be taken as broad indicator
of Market as whole. Similarly to measure and evaluate the performance of
various Sectors Sectoral Indices were created to capture the sentiment of
particular sector. This also enables to study about the overall performance of
market and sector in terms of return and risk
REVIEW OF LITERATURE:
There have been studies to study the
sectoral risk and return. Sectoral
Indices indirectly acts as measure of the performance of economic activity and
acts as barometer of sector performance. Demirer and Lien (2005)3
studied the correlation in different sectors in both upward and downward
direction in which it was observed that when the market is bullish there is
high correlation in Chinese market whereas in US Market the sectoral indices
showed strong correlation in bearish market and degree is stronger in extreme
movements. Dingmu Cao, Wen Long, Winning Yang(2013)2 found that
there have been stages when the markets are highly volatile then various
indices also tend to follow same trend however there are stages when there are
low correlation and many sector indices follow cyclical pattern of economy. Poshakwale
Sunil (2002)9 reported that daily returns from the stock indices in
India not strongly confirm to random walk. Swarna Lakshmi P (2013)8 on
studying the volatility across various sectors fond realty having highest
volatility. Gupta H (2016)7 studied the relationship of return on
investment of NIFTY with respect to the P/E value Shanmugasundram and Benedict
(2013)12 did a study on the volatility of the sectoral indices with
reference to Nifty in which they concluded that there do not exist any
significant difference across the risk of various indices and NIFTY. Prabahar,
Dhinakaran and Pandian (2008)10 concluded that systematic risk were
lesser in IT stocks than Unsystematic risk. Rajamohan, S., Muthukamu, M.,
(2014)11 studied about the risk return performance of the sectoral
indices of NSE and it was observed that there is a positive correlation of
return banking sector with other sectors. Most of the studies have been focused
on identifying factors which have impacted the broad market depicted by
diversified index however the current paper has tried to identify the impact of
various sector performances on market as a whole and also how sector
performance impacted each other in Indian market
SOURCE OF DATA:
Sensex data has been captured from
April 2001 to April 2016 and also data of 11 sectors have been taken from
different indices of Bombay Stock Exchange (Table 1).
Table 1
|
Indices
|
Date from where Data is
taken
|
Index Inception date
|
|
S& P Sensex
|
1-Apr-01
|
1-Jan-86
|
|
S& P BSE Auto Index
|
1-Apr-01
|
23-Aug-04
|
|
S&P BSE Capital Goods
|
1-Apr-01
|
1-Feb-99
|
|
S&P BSE Consumer Durable
|
1-Apr-01
|
9-Aug-99
|
|
S&P BSE FMCG
|
1-Apr-01
|
1-Apr-15
|
|
S&P BSE Health Care
|
1-Apr-01
|
1-Apr-15
|
|
S&P BSE IT
|
1-Apr-01
|
1-Apr-15
|
|
S&P BSE Metal
|
1-Apr-01
|
23-Aug-04
|
|
S&P BSE Oil &Gas
|
1-Apr-01
|
23-Aug-04
|
|
S&P BSE Power
|
1-Jan-05
|
9-Nov-07
|
|
S&P BSE Bankex
|
1-Jan-02
|
23-Jun-03
|
|
S&P BSE Realty
|
1-Jan-06
|
1-Jul-07
|
These indices in Bombay Stock
exchange are maintained by Asia Index Pvt. Ltd which is a joint venture between
Bombay Stock Exchange and Standard Poor rating agency. The company has also
done back testing and data prior to launch is not actual data but back tested
performance.
OBJECTIVE OF THE STUDY:
1) To study the performance of various
sectors in terms of risk and return
2) To study the relationship in returns
between various sectors over different horizon of investment period
3) To study the Granger relationship in
movement of sector indices
DATA ANALYSIS:
Compounded Returns were calculated for different holding
period which is monthly, One Year, Three Year, Five Year and Seven year for all
Indices and various tests were applied on it. To test the Normality of data for
return Jarque Bera Test was applied
n = no. of observations
k = no. of variables
S = Skewness
C = Kurtosis
On testing the Normality of Returns
by applying JB Test across different holding period it was observed (Table 2)
that for most of the sectors the returns are not normal for monthly investment
only FMCG sector was showing normal returns. Whereas when horizon was for one
year return Banking and Realty sector followed Normal return On testing
Normality for investment period of Three Years only FMCG and IT confirmed the
hypothesis of returns being normal. For testing investment period for five
years and more there were very few sectors whose return were normal. Hence to
check whether there exist substantial variability in returns across Sectors Non
parametric Test were used and for that as seen from Table 3 Kruscal Wallis T test
was applied and as observed there was no significant difference across returns
of various Sector and Sensex when Investment horizon is for Monthly however
other than that it showed there does exist substantial difference in returns
across various sectors Investment has to be evaluated from broadly two
perspective which is Return and Risk associated in it and both the factors have
to be seen jointly. Some common measurements which incorporate risk and return
are Sharp ratio and Treynor’s ratio however in this another common statistical
measurement Coefficient of Variation (CV) was used and higher CV means a better
reward and risk ratio.
Unit Root Test:
For testing this daily
time series data of various indices for the same period was taken and tests
were conducted through E-views 8. Granger and New bold (1974)5
observed results from the VAR models with non-stationary series will be
spurious. To perform Granger causality it was necessary to check whether all
the time series variables are stationary and have no Unit Root Problem. The
augmented Dicky–Fuller test (Dickey, Bell and Miller, 1986)4 was
used to examine whether the series are stationary or not.
All the series of Indices were non stationary at level
however on taking their first difference the series were converted into
stationary series
Granger Causality Test:
Variables are supposed to be Granger
cause over each other if variable X over a period of time influences another
variable such as Z if it is observed that the prediction of Value Z is based on
its own past value and also based on past value of X and predictions are better
than predicting based on past vale of Variable Y itself Granger causality was
tested on first difference series where Null Hypothesis was that there was no
Granger Causality between both variables. Before testing for Granger Causality
suitable lag order was obtained by Akaike
information criterion which indicated that lag
order of two as seen from Table 6. Three type of relationship was identified
between sectors which were Univariate where only one sector or indices has lead
or lagged relationship with other. Bivariate relationship where both indices
were influencing each other and thirdly some sectors showed no relationship
over each other movement. The nature of relationship was identified by F-value
and its significance at 0.05 level was observed by corresponding p-value On
testing for Granger Causality between sectors and Market for two lags in first
difference return Banking was observed having influence on all sectors other
than IT. Similarly Health Sector was also Granger causing all the sector except
Realty. Market (Sensex) was influencing almost all the sector other than FMCG.
Market has shown Bivariate Relationship with Health and Metal Indices. Movement
in IT Sector has not significantly influence to any other sector movement and
Market. Banking Sector depicted bivariate relationship with four sectors and
univariate leading indicator for six sectors
FINDINGS
OF STUDY:
1)
When the investment is monthly
there is no substantial difference in return across various sectors however
when the investment horizon is more than year then there exists difference
across various returns. In terms of risk reward ratio FMCG and Health care
Sector are the best performer whereas Realty and Power have the worst risk
reward ratio
2)
There exist high degree of
correlation among returns across various sectors however when the investment
horizon is for more than five years there is negative correlation of FMCG with
Power and Metal sector
3)
FMCG and Health Sectors are the
most defensive sectors and ideal for conservative investors and Capital Goods ,
Metal and Realty are aggressive and most sensitive sectors
4)
IT Sector is least dependent on
Overall market Performance and thus variations in this sector is largely
because of other Non-Systematic factors
5)
Returns in Banking Sectors are
largely explained by market conditions and sector performance is primarily
because of systematic risk
6)
Movement of FMCG Index is not
influenced by any other Sector Indices or market as a whole
7)
Movement in Banking sector are
highly influenced by other sectors and is also influencing movement in other
sectors
CONCLUSION:
Study
has tried to decipher relationship between returns of various sectoral indices
and also of market. It was observed that certain sectors such as Auto, Capital
goods and Bank have delivered better return than market(Sensex) as a whole
however in terms of Reward risk ratio FMCG and Health Sector have shown a
better risk and reward ratio. Both these sectors are also less influence by
changes in other sectors. This can be an indicator for investor building
portfolio can think of exposure in these sectors for getting some safe returns.
REFERENCES:
1.
Bekaert, G., Hodrick, R., and
Zhang, X. (2009). International stocks return co-movements. The Journal of
Finance, 64(6), 2591–2626.
2.
Cao, D., Long, W.,
and Yang, W. (2013). Sector Indices Correlation Analysis in China's Stock
Market. Procedia Computer Science, 17,
1241-1249.
3.
Demirer, R., and Lien, D. (2005).
Correlation and return dispersion dynamics in Chinese markets. International
Review of Financial Analysis, 14(4), 477-491.
4.
Dickey, D.A., Bell, W.R., and
Miller, R.B. (1986). Unit roots in time series models: Tests and implications. The
American Statistician, 40(1), 12–26.
5.
Granger, C.W.J. (1969).
Investigating Causal Relations by Econometric Models and Cross Spectral
Methods.Econometrica.37:424-35.
6.
Gupta, H. (2015).
A study on performance of Sensex and Evaluation of Investing Lumpsum or Monthly
Regular Investment in Equity on Risk and Return for Investor , International Journal of Development Research Vol. 5,
Issue, 04, pp. 4323-4327, April, 2015
7.
Gupta, H. (2016).
A Study on Evaluating P/E and its Relationship with the Return for NIFTY. International Journal of Innovative
Research and Development|| ISSN 2278–0211, 5(7).
8.
P. S. Lakshmi,
“Volatility patterns in various sectoral indices in Indian stock market,” Global
Journal of Management and Business Studies, vol. 3, no. 8, pp. 879-886,
2013.
9.
Poshakwale, S.
(2002). The random walk hypothesis in the emerging Indian stock market. Journal of Business Finance and
Accounting, 29(9‐10), 1275-1299.
10.
Prabahar,
R., Dhinakaran, J., and Pandian, P. (2008). Return and risk analysis of indian
information technology sector stocks. The
ICFAI Journal of Financial Risk Management, 5, 41-49.
11.
S. Rajamohan and
M. Muthukamu, “Bank nifty index and other sectoral indices of NSE-A comparative
study,” PARIPEX-Indian Journal of Research, vol. 3, no. 4, pp. 147-149,
2014.
12.
Shanmugasundaram
D., and Benedict, D. J. (2013). Volatility of the Indian sectoral indices–A
study with reference to National Stock Exchange. International Journal of Marketing,
Financial Services and Management Research, 2(8), 1-11.
=0.38) was least thus most of the
variation in IT sector is may be because of other factors related to IT sector
and for Banking (
) and Capital Goods (
the value is highest which indicates
that Banking and Capital goods sector is highly influenced by market. Similar
results were also visible when investment horizon increases and inference can
be drawn that variation in Banking and Capital Goods Sector returns are largely
explained by market variations whereas variations in IT, FMCG and Health care
sector have low relationship with Market Variations.