Empirical Evidence of the Dynamic Relationship between Stock

Price and Exchange Rate in India

Submitted by
MD AKRAM SABA
Email: akram.iitkgp@gmail.com

Under the Guidance of

PROF. BHAGIRATH BEHERA
Email: bhagirath@hss.iitkgp.ernet.in

Department of Humanities and Social Science

Indian Institute of Technology Kharagpur
Kharagpur, India 721302

DECLARATION
I certify that

a. The work contained in this report is original and has been done by me under the
guidance of my supervisor(s).
b. The work has not been submitted to any other Institute for any degree or diploma.
c. I have followed the guidelines provided by the Institute in preparing the report.
d. I have conformed to the norms and guidelines given in the Ethical Code of
Conduct of the Institute.
e. Whenever I have used materials (data, theoretical analysis, figures, and text) from
other sources, I have given due credit to them by citing them in the text of the
report and giving their details in the references. Further, I have taken permission
from the copyright owners of the sources, whenever necessary.

MD AKRAM SABA
(09HS2027)
5th Year Undergraduate Student
Department of Humanities and Social Science
Indian Institute of Technology Kharagpur
8th May 2014

Department of Humanities and Social Science

Indian Institute of Technology Kharagpur
Kharagpur, India 721302

Certificate

This is to certify that the Dissertation Report entitled, Empirical evidence of the dynamic
relationship between Stock price and exchange rate in India submitted by MD AKRAM
SABA to Indian Institute of Technology, Kharagpur, India, is a record of bonafide Project
work carried out by him under my supervision and guidance and is worthy of consideration as
his M.SC project in the Department of Humanities and Social Science, Indian Institute of
Technology, Kharagpur.

PROF. BHAGIRATH BEHERA

Department of Humanities and Social Science
Indian Institute of Technology Kharagpur
8th May 2014

ACKNOWLEDGEMENTS
I consider the completion of this research as dedication and support of a group of people rather
than my individual effort. I wish to express my gratitude to everyone who assisted me in
completing this research project.
First of all, I would like to express my sincere thanks and appreciation to my academic advisor
Prof. Bhagirath Behera for his support, guidance, encouragement, and help throughout the
M.Sc. Project.
I would also like to thank Prof. G.S.Hiremath for their constant support and encouragement
throughout my course.
I extend my heartfelt grateful to all who stood by my side in difficult times and guided me, my
friends, classmates, research scholars, and above all the Almighty. They helped me in every
possible way all through the project, without their guidance and cooperation I could not have
finished the work.
Last but not the least, I would like to express my sincerest gratitude to my parents, and my
sisters and brother, for their never ending support. Their constant encouragement, patience,
sacrifice and moral support saw me through the finish.

Dedicated to My Beloved Parents

For their endless love, support and encouragement

ABSTRACT
The paper investigated the dynamic relationship between Stock Market and Foreign
Exchange Market in India. We have considered the daily & monthly closing values of BSE
Sensex and daily & monthly average nominal exchange rates of Indian Rupee in terms of US
Dollar from April 2003 to March 2014 to conduct the study. The Standard Granger Causality
model and Generalized Autoregressive Conditional Heteroskedascity (GARCH) model were
used in establishing the relationship between domestic exchange rate and stock market.
Empirical result of Granger Causality test shows that stock market uni-directionally causes
exchange rate in an inverse relationship which tells that depreciation in the domestic currency
leads to an increase in stock market price in India. The outcome of ARCH-GARCH test
suggests that stock market return volatility and exchange rate return volatility are influenced
not only by their own ARCH and GARCH factors or internal factors but also by external
factors. The impact of the exchange rate return volatility on the stock return volatility is more
than 1, which means that little fluctuation in the exchange rate return will spill over more than
proportionately on the stock return movement for the longer period. In case of exchange rate
return as dependent variable, both internal shocks and external shock affect significantly on the
exchange rate return volatility. There is an inverse relationship between stock return volatility
and exchange rate return volatility which means stock return volatility dampens the exchange
rate return volatility for the short period.

Keywords: Stationarity, ADF, Phillips-Perron, Granger Causality, Volatility, ARCH, GARCH

Table of Contents
ACKNOWLEDGEMENTS .................................................................................................................... 4
ABSTRACT............................................................................................................................................ 6
1.

Introduction ..................................................................................................................................... 8

2.

Literature Review............................................................................................................................ 9

3. Objective of the Study ...................................................................................................................... 13

4. Data and Methodology ...................................................................................................................... 13
4.1 Data and Source ........................................................................................................................ 13
4.2 Methodology .............................................................................................................................. 14
4.2.1. Unit root tests .................................................................................................................... 15
4.2.2 Measurements and Characteristics of Volatility ............................................................. 18
5. Empirical Results .............................................................................................................................. 20
5.1 Cointegration and Granger Causality tests ............................................................................ 20
5.2 Estimating the volatility and the factors affecting the volatility ........................................... 22
6. Conclusion ........................................................................................................................................ 26
References ............................................................................................................................................. 29

1. Introduction
The aim of the study is to provide an empirical analysis of the interaction between the
exchange rate fluctuations and stock prices movement in Indian market. Three events Asian
financial crises of 1997-98, the advent of floating exchange rate in the early 1970s and financial
market reforms in the early 1990s in the emerging markets have prompted researchers into
determining the link between the stock market and foreign exchange market (Mishra, 2004).
The Asian crisis of 1997-98 has made a strong pitch for dynamic linkage between stock
prices and exchange rates. During the crisis period, the world has noticed how the emerging
markets collapsed due to substantial depreciation of exchange rates (in terms of USD) and
dramatic fall in the stock prices were closely related.
The liberalization of foreign capital controls and adoption of floating exchange rate
regime in India in early 1990s have widened the scope of studying the relationship between
exchange rates and stock prices. Liberalization of foreign capital controls has opened the
possibility of international investment in India and the adoption of floating exchange rate
regime has increased the volatility of foreign exchange market. Thus, detecting the association
between stock prices and exchange rates has become crucial for the academicians, practitioners
and policy makers.
However a survey of the available literature reveals divergent views of researchers on
the issue of whether foreign exchange rate variability influences stock market volatility (Frank
and Young, 1972; Solnik, 1987; Taylor and Tonks, 1989). The empirical debate regarding the
interaction between stock prices and exchange rates has been started few decades ago. Since
then a good number of empirical studies so far have been conducted to investigate the
relationship between the variables. But the researchers have found contradictory results
regarding the existence of relationship and the direction of relationship which has made the
area disconcerted environs of finance literature. So there is no empirical harmony among the
researchers regarding the interactions between stock prices and exchange rates which justify
the need of more research in this area to contribute to the literature.

Like other Asian emerging economies, Indian equity market has continued to grow and
has seen the relaxation of foreign investment restrictions primarily through country
deregulation. During the 1990s, India has initiated the financial sector reforms by way of

adopting international practices in its financial market. Over the years, Indian Rupee is slowly
moving towards full convertibility, which has also had an impact in the Indian capital market.
The two-way fungibility of ADRs/GDRs allowed by RBI has also possibly strengthened the
linkages between the stock and foreign exchange markets in India. In this background, this
paper aims at examining the dynamic linkages between foreign exchange of Indian Rupee in
term of US Dollar and BSE stock market price index in India using standard Granger causality
methodology and ARCH-GARCH model to measure the volatility of the variables and the
factors affecting the volatility. The rest of the paper has been arranged as follows: Section 2
gives the literature review, section 3 gives the objective of the study and Section 4 talks about
the data and methodology. Section 5 gives the empirical results. Finally Section 6 concludes.

2. Literature Review
Theoretical links between stock prices and exchange rates are explained by two
portfolio models. Firstly, Dornbusch and Fischer (1980) and Gavin (1989) suggested the
Flow-Oriented models of exchange rates, which posit that changes in exchange rates affect
the international competitiveness and trade balance of a firm as fluctuations in exchange rate
affects the value of the earnings and cost of its funds as many companies borrow in foreign
currencies to fund their operations and hence its stock price. A depreciation of the local
currency makes exporting goods attractive and leads to an increase in foreign demand and
hence revenue for the firm and its value would appreciate and hence the stock prices. On the
other hand, an appreciation of the local currency decreases profits for an exporting firm because
it leads to a decrease in foreign demand of its products. However, the sensitivity of the value
of an importing firm to exchange rate changes is just the opposite to that of an exporting firm.

Gavin (1989) argues that the innovations in the stock market, on the other hand, affect
aggregate demand through wealth, liquidity effects and indirectly the exchange rate.
Specifically a reduction in stock prices reduces wealth of local investors and further reduces
liquidity in the economy. The reduction in liquidity also reduces interest rates which in turn
induce capital outflows and in turn causes currency depreciation. Stock prices, generally
interpreted as the present values of future cash flows of firms, react to exchange rate changes
and form the link among future income, interest rate innovations, current investment and
consumption decisions. In addition, variations in exchange rates affect a firm's transaction
exposure. That is, exchange rate movements also affect the value of a firms future payables

(or receivables) denominated in foreign currency. Therefore, on a macro basis, the impact of
exchange rate fluctuations on stock market seems to depend on both the importance of a
countrys international trades in its economy and the degree of the trade imbalance.
Secondly, Branson (1983) and Frankel (1983) present the Stock-Oriented models of
exchange rates, which view exchange rates as equating the supply and demand for assets
(stocks and bonds). This approach determines exchange rate dynamics by giving the capital
account an important role. A blooming stock market would attract capital flows from foreign
investors, which may cause an increase in the demand for a countrys currency. The reverse
would happen in case of falling stock prices where the investors would try to sell their stocks
to avoid further losses and would convert their money into foreign currency to move out of the
country. There would be demand for foreign currency in exchange of local currency and it
would lead depreciation of local currency. As a result, rising (declining) stock prices would
lead to an appreciation (depreciation) in exchange rates. Since the values of financial assets are
determined by the present values of their future cash flows, expectations of relative currency
values play a considerable role in their price movements. Therefore, stock price innovations
may affect, or be affected by, exchange rate dynamics.

According to flow oriented model, when exchange rate depreciates (appreciates) in the
economy, share price of the domestic firm increases (decreases) through value of the earnings
and trade balances. Hence, we get a positive relationship between exchange rate (INR/USD)
and stock price when causality runs from former to latter. On the other hand, Stock oriented
model gives an inverse relationship between stock price and exchange rate (INR/USD) through
wealth and liquidity effect and interest rate adjustment in the economy.

The dynamic relationship between stock prices and exchange rates is of great interest
to many financial academicians and professionals, both for theoretical and empirical reasons,
since they both play a crucial role in influencing the development of a countrys economy.
Nonetheless, results are somewhat mixed as to whether stock indexes cause exchange rates or
vice versa and whether feedback effects (bi-causality) even exists among these financial
variables.
An early attempt to examine the exchange rate and stock price dynamics was by Franck
and Young (1972) who showed that there is no significant interaction between the variables.

Aggarwal (1981) made a study to find the relationship between exchange rates of US
dollar and changes in the indices of US stock prices and found a positive correlation.
Giovannini and Jorion (1987) also considered the exchange rates and stock prices of USA and
supported Aggarwal (1981).

Soenen and Hennigar (1988) studied the same market but considered a different time
period and contrast with prior studies by showing a significant negative relationship between
stock prices and exchange rates. Solnik (1987) made a slightly different study and tried to
detect the impact of several economic variables including the exchange rates on stock prices.
He concluded that changes in exchange rates do not have any significant impact over stock
prices. Jorion (1990) did a similar study to show the relationship between stock returns of US
multinational companies and the effective exchange rate of US dollar and found a moderate
relationship between the variables.
In a study by Mao and Kao (1990), exporting firms stock values were seen to be more
sensitive to changes in foreign exchange rates. Their findings also revealed another topical
issue of the relationship between stock prices at the macro and micro levels. Although theories
suggest causal relationship between exchange rate and stock prices, existing evidence indicates
a weak link between them at a micro level. On the macro level, Ma & Kao (1990) found that a
currency appreciation negatively affects the domestic stock market for an export-dominant
country and positively affects the domestic stock market for an import-dominant country,
which seems to be consistent with goods market theory.

Bahmani-Oskooee and Sohrabian (1992) used monthly values of S&P 500 index and
US dollar effective exchange rate for the period of 1973-88 and used cointegration and Granger
causality test to detect the relationship between the variables. They found no long run
relationship among these variables, but a dual causal relationship in the short run using Granger
(1969) causality tests. Amihud (1994) and Bartov and Bodnar (1994) found that lagged, and
not contemporaneous, changes in US dollar exchange rates, explain firm current stock returns.
Roll (1992) also studied the US stock prices and exchange rates and found a positive
relationship between the two markets. On the other hand, Chow etal. (1997) examined the same
markets but found no relationship between stock returns and real exchange rate returns. They

repeated the exercise with a longer time horizons and found a positive relationship between the
two variables.

Ajayi and Mougoue (1996) showed a negative short-run and positive long-run impact
of stock prices on domestic currency value. Yu (1997) studied Hong Kong, Tokyo and
Singapore markets by using daily data for a period of 1983-94. They traced bidirectional
relationship in Tokyo, no causation in the Singapore markets and also found that changes in
exchange rates Granger cause changes in stock prices.

Abdalla and Murinde (1997) employed co-integration test to examine the relationship
between stock prices and exchange rates for four Asian countries named as India, Pakistan,
South Korea and Philippines for a period of 1985 to 1994. They detected unidirectional
causality from exchange rates to stock prices for India, South Korea and Pakistan and found
causality runs from the opposite direction for Philippines.

Ajayi et al. (1998) studied markets of some advanced economies such as USA and
Korea and emerging economies such as Malaysia. They found out that there is a unidirectional
causality from the stock prices to foreign exchange markets in case of USA and Korea and no
relationship between the variables in case of Malaysia.

Mansor (2000) investigated Malaysian markets and found no long-run relationship

between stock prices and exchange rates, but he found a short-run causal relationship from
stock prices to exchange rates in bivariate cases. He also found a bi-directional causality in
some multivariate models. Wu (2000) did a similar study using stock prices and exchange rate.

Nieh and Lee (2001) found no significant long run relationship between stock prices
and exchange rates in G-7 countries, using both the Engle-Granger and Johansens
cointegration tests. Furthermore, they found ambiguous, and significant, short run relationships
for these countries. Nonetheless, in some countries, both stock indexes and exchange rates mat
serve to forecast the future paths of these variables. For example, they found currency
depreciation stimulates Canadian and UK stock markets with a one day lag, and that increases
in stock prices cause currency depreciation in Italy and Japan, again with a one day lag.

Engle and Rangel (2005) also examine the link between the unconditional volatility and
a number of macroeconomic variables. Bercker and Clement (2005) extended the SPLINE
GARCH model proposed by Engle and Rangel (2005) when they modelled stock market
volatility conditional on macroeconomic conditions. They incorporate macroeconomic
information directly into the estimation of such GARCH models. It was demonstrated that
forecasts of macroeconomic variables can be easily incorporated into volatility forecasts for
share index returns. Thus their model can lead to significantly different forecasts than
traditional GARCH type volatility models.

It is evident that the standard Granger causality method has been the most predominant
model used in most studies. Some researchers have used Granger Causality methodology on
the issue of stock market and exchange rate relationship. We will be finding the dynamic
relationship between these two variables using the same methodology in the recent years. In
general, empirical findings suggest the exchange rates and stock prices show positive
relationship in most countries. However, many studies have found that these variables have
predictive ability for each other, although the direction of causality seems to depend on
specific characteristics of the country analyzed.

3. Objective of the Study

There are two primary objectives of the study.

The paper explores the dynamic relationships between the stock prices and the exchange
rates using Johansen (1995) maximum likelihood cointegration test and Granger causality
test (1988) for the period of 11 years (April 2003- March 2014).

The paper also empirically analyzes the dynamic relationship between stock prices and
exchange rates based on autoregressive conditional heteroskedasticity (ARCH) and
multivariate generalized autoregressive conditional heteroskedasticity (GARCH) models
using daily data from 1st April, 2003 to 28th March 2014.

4. Data and Methodology

4.1 Data and Source
The data used in this study includes the monthly closing values of BSE Sensex and
monthly average nominal exchange rates of Indian Rupee in terms of US Dollar from April
2003 to March 2014 to find the causality relationship between these two variables. We will

denote the BSE Sensex and exchange rate as SP and ER respectively. To smoothen the time
series data, we transform both the data series into natural log form. The BSE Sensex data is
obtained from the Ace Equity Database and the exchange rate data is obtained from RBI
Database on Indian Economy. The data is also obtained on the daily basis for finding the
volatility of the time series using ARCH-GARCH model.

4.2 Methodology
The data series we use in the study are time series data. It is assumed that majority of
time series variables are non-stationary at level form or integrated of order 1 (Engle and
Granger, 1987). Using non stationary time series in a regression analysis may result in spurious
regression which was firstly pointed out by Granger and Newbold (1974). Thus before
analyzing time series data in an empirical study we should make stationarity test which is
commonly done by unit root test. There are a variety of unit root tests used in econometric
literature principally Augmented Dickey-Fuller (ADF) test and Phillip-Perron (PP) test. In this
study we use both unit root test to investigate whether the Yt time series data used in this study
are stationary or not.

Before applying unit root tests, it is suggested that optimal lag length should be chosen,
aiming at minimizing the residual sum of squares (RSS). Here, we select the optimal lag for
ADF based on its minimum Akaike information criteria (AIC) value.
For any statistical model, the AIC value is given as:

Ln( AIC )

2k
RSS
Ln(
)
n
n

(1)

Where k is the number of regressors (including the intercept) and n is the number of
observations.

For Phillips-Perron unit root test, we use Newey-West formula to obtain the optimal
lag length.

2
N 9

P Integer (4
)
100
Where N= Number of the observations,

P= Optimal lag

4.2.1. Unit root tests

Augmented Dickey-Fuller (1979) test is obtained by the following regression
m

Yt 1 2t Yt 1 i Yt 1 t

.............(2)

i 1

Where is the difference operator, , and are the coefficients to be estimated, Yt is the
variable whose time series properties are examined and , the white-noise error term.
Phillips and Perron (1988) test suggests a non-parametric method of controlling for
higher order autocorrelation in a series and is based on the following first order auto-regressive
AR (1) process:

Yt Yt 1 t

.............(3)

Where is the difference operator, is the constant, is the slope and Yt-1 is the first
lag of the variable Yt , whose time series properties are examined and , the white-noise error
term.
If the series used in the study found out to be integrated of the same order, it is useful
to test for co-integrating relationship between the integrated variables. For this purpose we
employ the Johansen procedure (Johansen, 1988; Johansen and Juselius, 1990) to test for the
possibility of a cointegrating relationship.
The Johansen method of cointegration applies maximum likelihood procedure to
determine the presence of cointegrating vectors in non-stationary time series as a vector
autoregressive (VAR):

Yt C

p 1

i Yt 1 t

.............(4)

i 1

Where Yt is a vector of non-stationary variables and C is the constant term. The

information on the coefficient matrix between the levels of the is decomposed as
where the relevant elements the matrix are adjustment coefficient and the matrix contains
cointegrating vectors. Johansen and Juselius (1990) specify two likelihood ratio test statistics
to test for the number of cointegrating vectors. The first likelihood ratio statistics for the null

hypothesis of exactly r cointegrating vectors against the alternative r+1 vectors is the maximum
eigenvalue statistic. The second statistic for the hypothesis of at most r cointegrating vectors
against the alternative is the trace statistic. Critical values for both test statistics are tabulated
in Johansen and Juselius (1990).
In the absence of any cointegrating relationship between the variables, the standard
Granger causality test base on Granger (1988) method will be applied. The Granger method
(Granger, 1988) seeks to determine how much of a variable, Y, can be explained by past values
of Y and whether adding lagged values of another variable, X, can improve the explanation.
The Granger method involves the estimation of the following equations:
p

SPt 0 1i SPt i 2i ERt i 1t

i 1

i 1

...........(5)

ERt 0 1i ERt i 2i SPt i 2t

i 1

...........(6)

i 1

in which SPt and ERt represent stock prices and exchange rates. 1t and 2t are uncorrelated
stationary random

process,

and t

denotes the time period. Failing to

reject

H 0 : 21 22 ....... 2 p 0 implies that exchange rates do not Granger cause stock prices.

On the other hand, failing to reject H 0 : 21 22 ......... 2 p 0 implies that stock prices
do not Granger cause exchange rates.
Based on the estimated OLS coecients for the equations (5) and (6) four dierent hypotheses
about the relationship between ER and SP can be formulated:
1. Unidirectional Granger-causality from ER to SP. In this case exchange rate increases the
prediction of the stock price but not vice versa.

i 1

i 1

2i 0 and 2i 0

2. Unidirectional Granger-causality from SP to ER. In this case the stock price increases the
p

prediction of the exchange rate but not vice versa. 2i 0 and

i 1

3. Bidirectional (or feedback) causality.

2i 0

i 1

and

2i 0

i 1

2i 0

i 1

4. Independence between SP and ER. In this case there is no Granger causality in any
p

direction.

2i 0

i 1

and

2i 0

i 1

To check the hypothesis of four possibilities, we apply the F test which is given as:

( RSS R RSS UR ) / p
RSS UR / n p

. (7)

which follows the F distribution with p and (n p) df.

If cointegration exists between SP and ER, the VECM is required in testing Granger
causality as shown below:
p

SPt 0 1i SPt i 2i ERt i 1 Z t 1 1t

i 1

i 1

ERt 0 1i ERt i 2i SPt i 1 Z t 1 2t

i 1

..........(8)

..........(9)

i 1

Where Zt-1 is the error correction term obtained from the cointegrating equation (4), so that
changes in the variables SPt and ERt are partly driven by the past values of Zt. The first
difference operator is marked by . The error correction coefficients, 1 and 1 , are expected
to capture the adjustments of SPt and ERt towards long-run equilibrium, whereas the
coefficients on SPt-i and ERt-I are expected to capture the short-run dynamics of the model.
Thus, in using equations (8) and (9) to test for the Granger-causal relationship between SPt and
ERt, we included the error-correction terms in order to introduce additional channels through
which causality could emerge and equilibrium could be re-established. Failing to reject
H 0 : 21 22 ....... 2 p 0 implies that exchange rates do not Granger cause stock prices

while failing to reject H 0 : 21 22 ......... 2 p 0 indicates stock prices do not Granger

cause exchange rates.

4.2.2 Measurements and Characteristics of Volatility

Volatility has been measured as standard deviation of the variables. The rates of returns
of stock price and exchange rate have been computed by taking a logarithmic difference of the
values of two successive periods. Symbolically, it may be stated as follows:

Rt Ln(

Xt
)
X t 1

...(10)

Where Xt is the value of the variable at time t.

Further, as discussed in the previous research works of Liu and Hung (2010) that, the
volatility of returns is categorized by a number of facts, such as volatility clusters, time-varying
volatility, and leptokurtic behaviour, but introduction of GARCH model of Bollerslev (1986)
and Engle (1982) has become a popular tool for modelling volatility.

Auto Regressive Conditional Heteroskedasticity (ARCH) models are used to

characterize and model observed time series. They are used whenever there is reason to believe
that, at any point in a series, the terms will have a characteristic size, or variance. In particular
ARCH models, we assume the variance of the current error term or innovation to be a function
of the actual sizes of the previous time periods' error terms: often the variance is related to the
squares of the previous errors. Engle (1982) suggests that the conditional variance H, can be
modelled as a function of the lagged U`s, i.e. the predictable volatility is dependent on past
news. The most detailed model developed was the pth order ARCH model which is shown as:

Ht W 1ut21 2ut22 .................. put2 p

(11)

Where W , 1 , 2 ,........... q parameters to be estimated and are considered positive. H t is the

conditional variance at period t. P is the number of lags included in the model and u t is the
innovation in return at time t.
An autoregressive moving average (ARMA) model is assumed for the error variance,
the model is a generalized autoregressive conditional heteroskedasticity (GARCH),
Bollerslev (1986)) model, which generalized the ARCH (p) model to GARCH (p, q) model,
such that:

H t W 1ut21 2ut22 .. put2 p

1H t 1 2 H t 2 ... q H t q
Where

(12)

W , 1 , 2 ,........... q , 1 , 2 ,............ p are parameters to be estimated and positive.

H t is the conditional variance at period t. P is the number of squared error term lags and q is
the number of conditional variances lags included in the model and u t is the innovation in
return at time t.
The GARCH (p, q) process defined above is stationary when (1 + 2 + .......... + q) +
( 1 + 2 + ............ + p) < 1.

To capture the volatility in the stock return and exchange rate return in the GARCH (1,1), we
assume a very simple mean equation as:
(13)

Yt ut

Where Yt represents the stock return or exchange rate return and Ut represents the random error
term. Now we obtain the residuals from the preceding regression and estimate the ARCH (1)
model, which is written as:

Ht W 1ut21 2ut22 .................. put2 p

(14)
If

there

is

no

autocorrelation

in

the

error

variance,

we

have

H 0 : 1 2 ......... p 0 in which case H t W , and we do not have the ARCH effect.

One can test the null hypothesis H0 by the usual F test, or alternatively, by computing nR2,
where R2 is the coefficient of determination from the above auxiliary regression. It can be
shown that nR2 2 , in large samples nR2 follows the chi-square distribution with degree of
freedom equal to the number of autoregressive terms in the above auxiliary regression.
If there is ARCH effect in the auxiliary equation, than we will use GARCH (1,1) model
to study the volatility of the daily return of the stock price and exchange rate. The GARCH
(1,1) can be written as:

Ht W ut21 Ht 1 Var ( X )t t

(15)

which says that the conditional variance of u at time t depends not only on the squared error
term in the previous time period [as in ARCH (1)] but also on its conditional variance in the
previous time period. To see the impact of volatility of the exogenous variable on the
conditional variance of u, we have included exogenous variable Xt in the model.

5. Empirical Results
5.1 Cointegration and Granger Causality tests
At first, we tested for the presence of unit roots tests and order of integration in the BSE
stock market index (SP) and exchange rate (ER) in the level form and, then in the first
difference. We used ADF test with constant and linear trend as suggested by Eangle and
Granger (1987) and PP test with no trend term. The lag length and bandwith for the unit root
tests were chosen on the basis of minimum AIC values to correct any serial correlation in the
residuals. The results of the unit root tests are given in Table 1.
Table 1: Unit root test results
ADF
Variables

Optimal lags#

Phillips Perron

Test statistics$

Newey-West

Test statistics@

lags
Ln (SP)

-2.748

-2.763

Ln (ER)

-1.808

-0.251

Ln (SP)

-10.285*

-10.210*

Ln (ER)

-7.308*

-7.937*

# Optimal lags is selected based on minimum Akaike Information Criteria (AIC)

$ Critical Values of ADF test statistics for 1%, 5% and 10% level of significance are: -4.030,-3.446, and -3.146
respectively.
@Critical Values of PP test statistics for 1%, 5% and 10% level of significance are: -3.500, -2.888, and
2.578respectively.
* Significant at 1% level.

Considering the results, it is clearly evident that null hypothesis H 0 : 0 of a unit

root test in the level form is accepted for both the variables as test statistics are lower than the
critical values in absolute term. We can say that exchange rates and stock prices at level form
are non-stationary data series. The PP test results also suggest the same. So, we took the first

difference of both the variables. After taking the first difference of both the time series and
applying the ADF and PP tests once again, this time results indicate that null hypothesis of a
unit root is rejected in both the cases. Therefore, the first difference of the data series of the
variables are stationary and hence we can say that data series are integrated of order one i.e. I
(1). After determining stationarity of the data series and order of integration, we progress to
cointegration test to find the presence of any cointegrating relationship between stock prices
and exchange rates. The results of cointegration test are given in Table-2.

Table 2: Multivariate Cointegration Test between Ln (SP) and Ln (ER)

Null

Alternate

Hypothesis#

Hypothesis

r=0

r=1

r1

r=2

Trace

5%

Max-

5%

Critical

Eigen

Critical

Values

Statistic

Values

11.8972*

15.41

11.8972

14.07

0.0000

3.76

0.0000

3.76

Variables

Ln(SP)/Ln(ER)

# r indicates number of cointegrating relationship.

* For a null hypothesis regarding r=0 means that there is no cointegrating relationship between SP and ER.
-

Optimal lag 2 is selected based on minimum AIC.

Trace Statistics is derived as Trace(r ) T

log(1 i) Where T is number of observations and

i r 1

is the r-th Eigen-value in ascending order.

Results clearly reveal that both trace test and maximum eigen value test accept the null
hypothesis of no cointegration i.e. H0: r=0. Thus we can say that there is no long term comovement between stock prices and exchange rates and none of the variables is predictable on
the basis of past values of other variable.
In the absence of any co-integrating relationship between the variables, we move to
standard Granger causality test to find out any causal relationship between stock prices and
exchange rates. To find out the causal relationship between the variables which are nonstationary at level form, the data series should be transformed into stationary ones (Oxley and
Greasley, 1998). Because it has been confirmed that Granger causality test is well specified if
they are applied in a standard vector autoregressive (VAR) form to differenced data for noncointegrated variables (MacDonaldand Kearney, 1987; Miller and Russek, 1990; Lyons and
Murinde 1994). Otherwise the inference from the F-statistics might be spurious because the

test statistics will have non-standard distributions. So we have transformed the level data series
into the first difference data series and used them for causality test. The results of Granger
Causality are given in Table 3.

Table 3: Granger causality test results

Null Hypothesis

F-Statistics

Probability

Lagged SP does not Granger Cause ER

6.4182*

0.0125

Lagged ER does not Granger Cause SP

0.91935

0.3395

* indicates significant causal relationship at 5%.

-

Optimal lag 1 is selected by minimum Akaike Information Criteria (AIC).

SPt 0.0147 0.0691SPt 1 0.3323ERt 1

P-value

(0.481)

(0.339)

ERt 0.0025 0.0653SPt 1 0.2187ERt 1

P-value

(0.013)

(0.018)

The results show that there is an uni-directional causal relationship from BSE stock
prices to exchange rates. So we can say that BSE stock prices influence exchange rates and
past values of stock prices can be used to improve the forecast of future exchange rates. There
is a negative relationship between stock price and exchange rate. The result is consistent with
the portfolio balance model that there is an inverse relationship between stock prices and
exchange rates. On the other hand, exchange rate does not granger cause stock price. So, we
can say that exchange rate does not affect BSE stock price and past values of exchange rate
cannot be used to forecast BSE stock price.
5.2 Estimating the volatility and the factors affecting the volatility
Figure 1 and Figure 2 show the pattern of daily returns of the BSE Sensex and
Exchange rate. Each of these series appears to show the signs of ARCH effect in that the
amplitude of the returns varies over time.

Figure 1: Volatility Clustering of Daily Returns of BSE Sensex

Figure 2: Volatility Clustering of Daily Returns of Exchange Rate

Both the figures reveal that, the returns continuously fluctuated around the mean
value that is close to zero for both the variables. The rate of return measures in both the cases
are in the positive and negative area. In case of stock return, it is clear that high volatility pattern
exist in the middle of the sample period. In case of exchange rate return, volatility has increased
substantially in the latter part of the sample period. Both the figures shows that high volatility
is followed by high volatility and low volatility are followed by low volatility. That means time
series have important time varying variances. Hence, it is appropriate to put conditional
variance into the function to clarify the impact of risk on the returns.

Table 4: Descriptive Statistics of Daily Returns

Basis
Observation Period

Stock Price Return

Exchange Rate Return

1st April 2003- 28thMarch 2014 1st April 2003- 28th March 2014

Number of Observation

2668

2668

Mean

0.0734

0.0089

Median

0.1170

0.0000

Maximum

15.9900

4.0200

Minimum

-11.8100

-3.0060

Standard Deviation

1.6102

0.5029

Variance

2.5929

0.2529

Skewness

-0.0435

0.2234

Kurtosis

10.8893

8.4644

Note: Skewness is a measure of asymmetry of the distribution of the time series around its mean.
Kurtosis measures the peakedness or flatness of the distribution of the series.

Table 4 presents some descriptive statistics for each of the daily returns. The maximum
and minimum values of the stock return are much higher as compared to the exchange rate
return. From the daily standard deviation, we see that the stock return is the more volatile than
exchange rate return. Since Skewness is less than zero for stock return, i.e., left skewed
distribution - most values are concentrated on the right of the mean, with extreme values to the
left. On the other hand, since Skewness is greater than zero for exchange rate return, i.e., right
skewed distribution - most values are concentrated on left of the mean, with extreme values to
the right. Both the series show the evidence of fat tails, since the kurtosis exceeds 3, which is
the normal value, i.e., Leptokurtic distribution, sharper than a normal distribution, with values
concentrated around the mean and thicker tails. This means high probability for extreme values.

Table 5: Descriptive statistic of ARCH-LM Test (at lag = 1) for Daily Returns.
Distribution

Chi-square

P>Chi-Square

Stock return

113.791*

0.000

Exchange rate return

140.912*

0.000

* indicates the rejection of null hypothesis of no ARCH effect in the residuals.

The existence of a leptokurtic distribution and presence of volatility clustering suggests

an ARCH or GARCH process. The Table 5 shows substantial evidence of ARCH effects in the
residuals as judged by ARCH-LM test. ARCH-LM test statistics is the Lagrange Multiplier test
statistic for the presence of ARCH effect. Under the null hypothesis of no heteroskedasticity,
it is distributed as a chi-square. Thus, the result gives the implication of the use of GARCH
model to estimate the volatility of the stock return and exchange rate.
Table 6: Results of ARCH and GARCH model
Dependent variable

Stock return

Exchange rate return

Constant (W)

-3.132*(0.000)

-6.733*(0.000)

ARCH ()

0.112*(0.000)

0.228*(0.000)

GARCH ()

0.867*(0.000)

0.798*(0.000)

Var (Xt)

1.063*(0.000)

-0.532*(0.000)

* indicates the significance of the coefficients in the GARCH (1,1)

Table-6 depicts the result of GARCH (1,1) estimation of stock return and exchange rate
return volatility. For stock return volatility as dependent variable, GARCH (1,1) model took
13 iterations to maximize the likelihood function under the normal distribution. The four
coefficients in the variance equation are listed as Constant(W), the intercept; ARCH(1), the
first lag of the squared error term; GARCH(1), the first lag of the conditional variance and
Variance (Xt), volatility of the exchange rate return. All the coefficients of the variables are
significant. The coefficient of ARCH term is positive and statistically significant, indicating
that previous days information on BSE stock return influences the present days volatility of
the stock return. The coefficient of GARCH term is also positive and statistically significant,
indicating that previous days volatility of stock return affects the present days volatility of
the stock return. The finding also indicate that the coefficient of the Var (ERRt) is also positive
and statistically significant. Higher the volatility in the exchange rate return (ERR) increases
the volatility of stock market return. This means that an increase in exchange rate volatility will
increase the present days volatility of stock return more than proportionately. Hence, we can
say that volatility of stock returns is not only caused by its own shock but also by the external
shock.

Similarly, for exchange rate return volatility as dependent variable, all the four
coefficients are statistically significant. From the results, we can say that the previous days

information on exchange rate return [as in ARCH], previous days volatility of exchange rate
return [as in GARCH] and stock return volatility (external shock) are influencing the present
days volatility of exchange rate return. However, the coefficient of Var (SRt) is negative,
indicating that that higher volatility in the exchange dampens stock market activity.

Moreover, the GARCH effect is higher than ARCH effect in both the cases. Notice that
the coefficients of ARCH and GARCH terms sum up to a number less than one, which is
required to have a mean reverting variance process. Since the sum is very close to one, this
process only mean reverts slowly. Also one can observe that, the two returns taken in this study
are highly correlated, and there is a considerable coexistence in their movements.

6. Conclusion
In this study we have explored the association between two important component of an
economy named as stock prices and exchange rates. We found that dynamic relationship
between these variables using Granger causality and ARCH-GARCH model. First of all, we
applied unit root test to find the stationarity of data series. The results show that both the data
series of the variables are non-stationary and integrated of order one. Then we applied Johansen
procedure to test for the possibility of a cointegrating relationship. Result shows that there is
no cointegrating relationship between stock prices and exchange rates. That means there is no
long-term co-movement between the variables and none of the variables is predictable on the
basis of past values of other variable. In the absence of any co-integrating relationship between
the variables, we move to standard Granger causality test to find out any causal relationship
between stock prices and exchange rates. Results shows that stock prices cause exchange rates
in an inverse relationship indicating that when stock price rises (falls), the exchange rate
appreciates (depreciates). This finding supports the result of stock oriented approach of
portfolio model. As the stock price of the domestic firms increase, foreign investors will like
to invest in the domestic firm. The demand of the Indian rupee will start increasing, hence local
currency appreciates and ultimately exchange rate appreciates. On the other hand, exchange
rates do not cause stock prices which means that exchange rate does not contribute in the
forecasting of the stock price in the economy. So there is an uni-directional causal relationship
between stock prices and exchange rates.
The fluctuations in the stock return and exchange rate return are closely related to one
another. Using ARCH-GARCH model, we observed that how volatility of the stock return and

exchange rate return are affected by own shocks and external shocks. The present days
volatility of the stock return is being influenced by past days information on stock return, past
days volatility of stock return and volatility of exchange rate return. The impact of the
exchange rate return volatility on the stock return volatility is more than 1, which means that
little fluctuation in the exchange rate return will spill over more than proportionately on the
stock return movement for the longer period.
Considering exchange rate return as dependent variable, both internal shocks (ARCH
and GARCH terms) and external shock (stock return volatility) affect significantly on the
exchange rate return volatility. There is an inverse relationship between stock return volatility
and exchange rate return volatility which means stock return volatility dampens the exchange
rate return volatility for the short period.
Exchange rate volatility has attracted much attention in financial economics in
developed and developing economies due to its implications in the financial markets, especially
the stock market. Different implications were observed between exchange rate volatility and
stock market returns depreciation in the local currency leads to increases in stock market
prices in the long run. Where as in the short run it reduces stock market returns. This is in
conformity with some empirical studies which propose that exchange rate depreciation is good
for stock markets especially where the stock market is located in a highly export driven
economy.
It was therefore recommended that for a better stock market performance, policy
makers should put in place measures that will ensure stable foreign exchange environment,
since any disturbances in the exchange rate environment may affect the stock markets
activities. So to attract investors (especially foreign direct investment) means that we should
have a stable exchange rate system. A volatile exchange rate could raise strategic and
managerial issues because it could lead to losses or gains. This may create uncertainty in
investors as to invest or not to invest in the market. Hence, to boost investor confidence, there
would be the need for policy makers intervention in times of abnormal volatility.
The volatile nature of the exchange rate market in the country also means that firms
that import raw materials or market their product internationally need to make use of forward
contracts in other to hedge their payables and receipts. This will enable them to lock in so as to
go round the problem of exchange rate volatility.

It is also recommended that investors could take into consideration the nature of
volatility in the exchange rate in the economy to make an informed decision as to where to
direct their investments. So that whenever the local currency depreciates, it is a signal that the
stock market returns is likely to appreciate; especially for an import dominated economy. But
this argument is based on an improvement in the international competitiveness of the local
firms.
Finally, it is suggested base on this study that other researchers can use data for other
countries and periods of time to study further the macroeconomic determinants of stock market
volatility in real time.

References
Abdalla, I. S. A. and V. Murinde (1997), Exchange Rate and Stock Price Interactions in
Emerging Financial Markets: Evidence on India, Korea, Pakistan, and Philippines, Applied
Financial Economics, 7, pp. 25-35.
Aggarwal, R (1981), Exchange rates and stock prices: A study of the US capital markets under
floating exchange rates, Akron Business and Economic Review, 12, pp. 7-12.
Ajayi, R. A., Friedman, J. & Mehdian, S. M (1998), On the relationship between stock returns
and exchange rates: Tests of Granger causality, Global Finance, 9, pp. 241-251.
Ajayi, R. A. and M. Mougoue (1996), On the Dynamic Relation between Stock Prices and
Exchange Rates, Journal of Financial Research, 19, pp. 193-207.
Bahmani-Oskooee, O. M. & Sohrabian, A (1992), Stock Prices and the effective exchange
rate of the dollar, Applied Economics, 24, pp. 459-464.
Bhattacharya, B. & Mukherjee, J (2003), Causal Relationship between Stock Market and
Exchange Rate, Foreign Exchange Reserves and Value of Trade Balance: A Case Study for
India, Paper presented at the Fifth Annual Conference on Money and Finance in the Indian
Economy, January 2003.
Bartov, E. and M. G. Bodnar (1994), Firm valuation, earning expectations and the Exchange
rate exposure effect, Journal of Finance, pp. 1755-1786.
Bollerslev, T (1986), Generalized Autoregressive Conditional Heteroskedasticity, Journal of
Econometrics, 31, pp.307-327.
Dornbusch, R. and S. Fischer (1980), Exchange Rates and Current Account, American
Economic Review, 70, pp. 960-971.
Engle, F. R (1982), Autoregressive Conditional Heteroskedasticity with Estimates of the
Variance of United Kingdom Inflation, Econometrica, Vol. 50, No. 4, pp. 987-1007.
Frank, P., and A. Young (1972), Stock Price Reaction of Multinational Firms to Exchange
Realignments, Financial Management, Winter, Principles of Economics, New York, McGraw
Hill/Irwin, pp.6673.
Gujarati D. N (2003), Basic Econometric., New York: The McGraw-Hill companies Inc.

Gavin, M (1989), The Stock Market and Exchange Rate Dynamics, Journal of International
Money and Finance, 8, 2, pp. 181-200.
Granger, C.W.J., B. Huang and C. Yang (2000), A Bivariate Causality between Stock Prices
and Exchange Rates: Evidence from Recent Asian flu, Quarterly Review of Economics and
Finance, 40 (3), pp.337-354.
Johansen, S (1991), Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian,
Vector Autoregressive Models, Econometrica 58, pp. 165-188.
Mansor H. I (2000), Cointegration and Granger causality tests of stock price and exchange
rate interactions in Malaysia, ASEAN Economic Bulletin, 17, pp. 36-47.
Mishra, K. A (2004), Stock Market and Foreign Exchange market in India: Are they related?,
South Asia Economic Journal, 5:2, Sage Publications, New Delhi.
Mao, C. K. G., and W. Kao (1990), On Exchange Rate Changes and Stock Price Reactions,
Journal of Business Finance and Accounting, 17 (3), pp.441-449. 10.
Nieh, C. C. & Lee, C. F (2001), Dynamic relationship between stock prices and exchange
rates for G-7 countries, The Quarterly Review of Economics and Finance, 41, pp. 477490.
Roll, R. (1992), Industrial structure and the comparative behavior of international stock
markets indices, Journal of Finance, 47, pp. 3-41.
Soenen, L. A. & Hennigar, E. S (1988), An analysis of exchange rates and stock prices: The
US experience between 1980 and 1986, Akron Business and Economic Review, Winter, pp.
7-16.
Solnik, B (1987), Using financial prices to test exchange rate models: A note, Journal of
Finance, 42, pp. 141-149.
Taylor, M.P., and I. Tonks (1989), The Internationalization of Stock Markets and the
Abolition of UK Exchange Control, Review of Economics and Statistics, 71, pp.33236.