Tải bản đầy đủ

The relationship between vietnam united states and related asian stock markets luận văn thạc sĩ

MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS HOCHIMINH CITY

TRẦN THÙY HUYÊN

THE RELATIONSHIP BETWEEN VIETNAM,
UNITED STATES AND RELATED A
ASIAN
STOCK MARKETS

MASTER THESIS

Ho Chi Minh City 2011


MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS HOCHIMINH CITY

TRẦN THÙY HUYÊN

THE RELATIONSHIP BETWEEN VIETNAM,

UNITED STATES AND RELATED ASIAN
STOCK MARKETS

MAJOR: BANKING AND FINANCE
MAJOR CODE: 60.31.12

MASTER THESIS

Supervisor: DOCTOR VO XUAN VINH
PROFESSOR TRAN HOANG NGAN
Ho Chi Minh City 2011


ACKNOWLEDGEMENT
First and foremost, I am very much indebted to Doctor Vo Xuan Vinh for the
sociable, motivation and professional guidance and comment. I understand you
sacrificed your valuable time and resources to help me conduct this thesis.
Secondly, my heartfelt gratitude goes to my supervisor the Professor Tran Hoang
Ngan. It is by his clemency that I am writing this thesis today. Thank you very
much for your understanding, commitment and your advice regarding my thesis.
Professors and the entire lecturers in University of economics HCMC, I thank you
for academic guidance and other support during my study.
My master’s classmates, I thank you for your advice and I will always cherish the
moments we had together.
Last but not least, I would like to thank my family, especially my husband, my baby
for the moral support and patience. I am particularly grateful to them for the moral
and advisory support during the entire of my studies and my life.

i


ABSTRACT
The international relationship amongst the stock markets plays an important roles
and implications for cost of capital and international portfolio diversification to both
Vietnamese and foreigner investors. Vietnam becomes widely opening the trade and
equity market since its entering to WTO in 2007. And recent movements of
international financial liberation in Vietnam and its region raise questions whether
Vietnam and its region stock markets in Asian are being integrated into world stock
markets. This research aims to examine that Vietnam and related Asian stock
markets have interrelationships to the world stock markets using the daily data for

the period 2005-2010. In determining the impact of the international financial crisis
on interrelationship amongst stock markets, the research period is being divided into
two sample periods: the pre crisis period spans from January 1st, 2005 to July 22,
2007 and the crisis period spans from July 23, 2007 to June 30, 2010. Seven indices
of six stock markets are chosen based on the countries’ level of development and
geographical factor. Six stock markets are Vietnam (Vietnam index), Singapore
(Straits times’ index), Hong Kong (Hang Seng index), China (Shanghai stock
exchange), Japan (Nikkei index) and United States (Standard & Poor’s index and
Dow Jones index).
The first content examines whether there is long run comovement amongst the
stock markets in sample. Bivariate and multivariate Johansen test are applied to
daily stock price and daily stock return after conducting the tests for unit root with
three tests of Dickey Fuller, Augmented Dickey Fuller and Phillip Perron. The
second content investigates whether there is causal relationship amongst these
markets using Granger causality test. The converging test is employed to know
whether they are converging over time. To analyses the return linkage, we replied
on variance decomposition analysis.
The first finding of research is that only two cointegrations exist between two pairs
of US-China and US-Hong Kong stock market in the pre crisis period and no
ii


cointegrations exist in the crisis period when we conduct the bivariate Johansen test
for daily stock price. Moreover, we also could not find the cointegration exist when
multivariate Johansen tests when daily stock price is used. These results imply that
the portfolio diversification is potentially worthwhile for investors. However, the
results of both bivariate and multivariate test for daily return are opposite.
Cointegrations are found in the pre crisis and during the crisis period. It can be
concluded that portfolio diversification may not bring benefits to investors. The
Granger causality test results indicate the current global financial crisis affect
significantly to the Granger causality relationship amongst the markets in sample
and US market are recognized as a significant influence on the returns of Asian
stock markets. In addition, converging trend test results find that seven indies of six
stock markets converge towards the group in whole period. The five Asian stock
markets are also converging to US stock market in the whole period and crisis
period. Except for Hong Kong and China, the other markets are converging to the
group and US stock markets in the pre crisis time. When apply the variance
decomposition analysis, we get the findings demonstrate that most of indices of six
stock markets in our sample demonstrate the strong endogenous characteristics
except for US, Vietnam and China stock market. Generally, the results of this
research find the relationship amongst Vietnam, US and other related Asian stock
markets in both sub periods. Hence, the Vietnam and international investors should
focus on portfolio diversification to other stock markets. On the other hand, policy
makers should draw up the appropriate macroeconomic strategies to promote
economic performance in Asian region and in the world because high international
financial integration will help to upgrade the financial development.

Key word: cointegration, stock markets, Granger causality, integration

iii


TABLE OF CONTENT
Acknowledgement ................................................................................................ i
Abstract ............................................................................................................... ii
TABLE OF CONTENT...................................................................................... iv
LIST OF FIGURES ............................................................................................ vi
LIST OF TABLE ............................................................................................... vi
I. INTRODUCTION.......................................................................................... 1
1. Rationales of research ...................................................................................... 1
2. Problem statements ......................................................................................... 3
3. Research objectives and research questions ..................................................... 4
3.1 Research objectives ....................................................................................... 4
3.2 Research questions ........................................................................................ 4
4. Scope and methodology ................................................................................... 5
4.1 Scope of research ........................................................................................... 5
4.2 Methodology of research ............................................................................... 5
5. Structure of research ........................................................................................ 7
II. LITERATURE REVIEW ............................................................................ 8
III. METHODOLOGY.................................................................................... 15
1. Johansen Cointegration test ........................................................................... 15
1.1 Unit root test ................................................................................................ 16
1.1.1 The Dickey-Fuller test (DF) ...................................................................... 16
1.1.2 The Augmented Dickey-Fuller test (ADF) ................................................ 17
1.1.3 The Phillips-Perron test (PP) ..................................................................... 18
1.2 Cointegration test (Johansen) ....................................................................... 19
2. Granger causality ........................................................................................... 20
3. Converging trend ........................................................................................... 22
4. Variance decomposition analysis ................................................................... 22
IV. DATA AND DATA DESCRIPTION ........................................................ 24
1.Data ................................................................................................................ 24
2. Data descriptive statistics .............................................................................. 26
2.1 Summary statistics ...................................................................................... 26
2.2 Correlation................................................................................................... 30
V. EMPIRICAL RESULTS ............................................................................ 33
1. Analysis of results of unit root test................................................................. 33
1.1 The Dickey-Fuller test (DF)......................................................................... 33
1.2 The Augmented Dickey-Fuller test (ADF) ................................................... 34
1.3 The Phillips-Perron test (PP)........................................................................ 35

iv


2. Cointegration ................................................................................................. 38
2.1 Johansen cointegration test with daily stock price ........................................ 38
2.1.1 Bi-variate cointegration test ...................................................................... 38
2.1.2 Multivariate cointegration test .................................................................. 44
2.2 Johansen cointegration test with daily stock returns ..................................... 45
2.2.1 Bi-variate cointegration test ...................................................................... 45
2.2.2 Multivariate cointegration test .................................................................. 50
3. Granger causality ........................................................................................... 55
4. Converging trend ........................................................................................... 59
5. Variance decomposition analysis ................................................................... 61
VI. CONCLUSION.......................................................................................... 64
1. Conclusion related to research questions ....................................................... 64
2. Implications ................................................................................................... 68
3. Contribution of research ................................................................................ 70
4. Limitation and Further research ..................................................................... 71
Appendix ........................................................................................................... 73
List of References ............................................................................................ 130

v


LIST OF FIGURES
Figure 1.1: Structure of research ......................................................................... 7
Figure 4.1: Movement of daily closing price of stock index from Jan 1st 2005 to
Jun 2010 ......................................................................................................... 25
Figure 5.1: Short run causality effects (whole period) ....................................... 56
Figure 5.2: Short run causality effects (pre crisis period) .................................. 57
Figure 5.3: Short run causality effects (crisis period) ........................................ 59

LIST OF TABLES
Table 4.1 Summary statistics of the stock returns (Whole period) ..................... 27
Table 4.2: Summary statistics of the stock returns (Pre-crisis period) ............... 28
Table 4.3: Summary statistics of the stock returns (Crisis period) ..................... 29
Table 4.4: Correlation matrix of average daily return (Whole period) ............... 30
Table 4.5: Correlation matrix of average daily return (Pre-crisis period)........... 31
Table 4.6: Correlation matrix of average daily return (Crisis period) ................ 31
Table 5.1: Var lag Order Selection ................................................................... 33
Table 5.2: DF unit root test (daily price) ........................................................... 34
Table 5.3: ADF unit root test (daily price) ........................................................ 35
Table 5.4: PP unit root test (daily price)............................................................ 36
Table 5.5: DF unit root test (daily return) ......................................................... 36
Table 5.6: ADF unit root test (daily return)....................................................... 37
Table 5.7: PP unit root test (daily return) .......................................................... 37
Table 5.8: Bi-variate cointegration test result with intercept (no trend) in
CE and test VAR (daily price of all stocks) ..................................................... 40
Table 5.9: Multivariate cointegration test result with intercept (no trend) in
CE and test VAR (daily stock price) ................................................................. 44
Table 5.10: Bi-variate cointegration test result with intercept (no trend) in CE
and test VAR .................................................................................................... 46
Table 5.11: Multivariate cointegration test result with intercept (no trend) in
CE and test VAR .............................................................................................. 51
Table 5.12: Granger causality results (Whole period) ....................................... 56
Table 5.13: Granger causality results (Pre crisis period) ................................... 57
Table 5.14: Granger causality results (Crisis period) ........................................ 58
Table 5.15: Test result for converging trend ..................................................... 60
Table 5.19: the comparision of degree of exogeneity in pre crisis and crisis
period ............................................................................................................... 63
Table 5.16: the comparision of degree of exogeneity in whole period ......... 72
Table 5.17: the comparision of degree of exogeneity in pre crisis period .... 75
Table 5.18: the comparision of degree of exogeneity in crisis period .......... 78

vi


I. INTRODUCTION
1. Rationales of research
It is well known that US is a big and main trading partner of many developed and
developing countries in Asian and in the world. As a result, anything happens to the
US economy will may cause effect to these Asian countries’ economy. This
interrelationship phenomenon in international market is a result of the liberalization
of capital markets in developed and developing countries and the increasing variety
and complexity of financial instruments. Financial specialists advocate that the US
stock market fluctuation will bring effects to many other stock markets in the world.
For example, the Nikkei 225 (Japan stock market) and Hangseng (Hong Kong stock
market) perpetually cling to the S&P 500 (US stock market). While some previous
researches in Vietnam believe that Vietnam stock market is frigid with US and
regional stock markets.
Researches on regional stock market linkages have become increasingly important
for most investors. The Asian region is vulnerable to ‘shocks’ (financial crisis as an
example) and where the crisis is contagious, it can affect the entire region. For this
reason, the countries in the region should become more concerned about their
interdependency in the event of any occurrences of any financial crisis. For
instance, the Asian financial crisis began with collapse of Thai baht in July 1997
and its stock market, and the subsequent erosion in Hong Kong and other Asian
markets in October 1997 and as a result, the co-movement amongst the Asian
financial markets increased (Maran, (2010). Thus, a question can be raised from the
2007-2010 international financial crisis whether this international financial crisis
causes the changes of interrelationship amongst Vietnam stock market, US and
region stock markets in short run and long run.

Page 2


In globalization integration economic environment, both investors and portfolio
managers shall pay more attention to the knowledge of the international stock
market structure. Many financial theories suggest that individual and institutional
investors should hold a well-diversified portfolio to reduce risk. As investors
become more risk averse, further risk diversification continues to be their main
concern. To the international investor who is willing to make portfolio investments
in different stock markets, they really need to know that diversification can give
some gain or not. The world co-movements amongst financial markets have
reduced the diversification benefit. The international financial markets are quickly
integrating into a global market since investors are driven to developing countries
searching for higher returns and opportunities for risk diversification. If the stock
markets amongst different countries move together, investment in different stock
markets shall not create any long term gain to portfolio diversification. Hence
researching results of stock market integration are useful when Asian economies are
fastest growing economies. This research results could be useful for investors,
portfolio managers, corporate executives and policy makers.
To our best knowledge, most empirical works have focused on developed markets,
developing markets of South East Asia such as Becker et al (1990), Mak (1992) or
Chan (1992). However, researches on the Vietnam stock market or Vietnam stock
market’s data of the 2007-2010 international financial crisis have been very limited.
This paper will investigates the interrelationship in long and short run amongst
Vietnam, US and other related Asian stock markets in the pre crisis in 2007-2010
and during the crisis of period 2007-2010. By this research, we hope to contribute
towards adding the literature by providing the latest empirical proof on this topic.

Page 3


2. Problem statement
Vietnam becomes WTO’s 150th member on 11 January 2007. According to the
route of entering WTO, Vietnam opens the economics including the financial
market as from the end of the year 2007. It means that Vietnam will play an
important role in world economics. Whatever happens to Vietnam economics may
affect to the economic in region and the world. Conversely, the regional and
international economics change will cause effect to Vietnam. And the stock market
is not considered an exception. However, there is an argument that some researchers
conclude Vietnam stock market is frigid with world stock market while other
researchers believe that Vietnam is affected by the US and some other stock
markets in region. In addition, it is well know that many investors in Vietnam stock
market still invest with a crowd psychology or “tam ly dam dong” in Vietnamese.
As a result, not many investors pay attention to the region or world stock market’s
fluctuation.
The issues of financial integration or stock markets cointegration between emerging
stock markets and developed stock markets have attracted a great deal of interest of
policy makers and finance researchers. The emerging stock markets in some
developing countries have achieved considerable improvements in recent time.
These improvements of emerging stock markets come from some factors such as:
stock market renovations, financial liberalization, economics policies… An
important point for this research is because there is an increase in funds flowing
from developed markets such as US and Japan toward developing markets like
Vietnam, Hong Kong, China… Therefore, these markets are becoming increasingly
important in terms of portfolio management Hawawini (1994).
In summary the problem to be addressed in this research is to study the relationship
amongst Vietnam, US and other Asian stock markets. The research results will be
useful to investors, policy makers and academic scholars.
Page 4


3. Research objectives and research questions
3.1 Research objectives
Research objectives are defined as the vision of researcher of the research problem.
Their roles are explanation the purpose of the research in measurable and defined
standards of what the research should accomplish Zikmund (1997). In order to solve
the research problems, this research intends to achieve the following objectives:
The first objective of this research is to examine the interrelationship amongst the
stock markets of Vietnam, US, Singapore, Hong Kong, China and Japan using the
daily data for the period from January 1st 2005 to June 30th 2010.
The second objective is to evaluate the level of stock markets integration and how
these stock markets affect together.
The third objective is to study the changes of integration through the 2007-2010
financial crisis. From there, investors can forecast movement of Vietnam stock
market replying on the world and region stock market’s movement. And the policy
makers can issue effective policies when the international crises occur.
Final objective is to suggest investment and policy advice.
3.2 Research questions
According to Zikmund (1997), research questions involve the research translation of
problem into the need for inquiry. As above research problems and research
objectives, the tasks of this research are to answer questions following:
Question 1: Is there a relationship between Vietnam and Asian stock market with
US stock market? (Answer in part VI)

Page 5


Question 2: Is there a relationship between Vietnam stock market & related Asian
stock markets of Singapore, Hong Kong, China and Japan? (Answer in part VI)
Question 3: Is there a cause-effect relationship between these stock markets? How
the relationship changes before and during the global financial crisis of 2007-2010?
(Answer in VI)

4. Scope and methodology
4.1 Scope of research
This research examines the relationship amongst Vietnam, US and the Vietnam’s
region stock markets. We choose some countries around Vietnam as a
representative of Asian stock markets such as Singapore, Hong Kong, China and
Japan. The changes of these stock markets relation through the international
financial crisis in 2007-2010 are conducted in this research so the period time is
chosen from 2005 to June 2010. Choosing more countries and longer period of time
may give more exact and convinced results.

4.2 Methodology of research
In order to solve the research problems, achieve the research objectives and answer
the research questions, it is quite important to choose and apply suitable
methodologies. Based on the theory in literature review and guideline from Brooks
(2008), we apply some following methodologies and econometric testings as a brief
description:
• Carrying out a review of the relevant theoretical literature and empirical
literature.

Page 6


• Getting daily data of each index from yahoo.finance website from 2005 to
June 2010.
• Conducting data descriptive statistics (summary basic statistic and simple
correlation estimation). Some of basic statistic tests are mean, median,
standard deviation, skewness, kurtosis and others.
• Applying the test for stationary with three tests: The Dickey-Fuller test (DF),
The Augmented Dickey-Fuller test (ADF) and The Phillips-Perron test (PP).
• Performing the Johansen cointegration test to test for cointegration amongstst
the sample stock markets.
• Carrying out the Granger causality test to find the causal effect amongstst
these markets.
• Executing the test of converging trend to identify the converging or
diverging trend of variables.
• Implementing variance decomposition analysis to identify exogenous and
endogenous variable.
Above methodologies are test by the Eview 6 software. Part 3 will mention detail of
each above methodologies.

Page 7


5. Structure of research
This thesis is not followed the conventional way of a normal thesis in Vietnam
because each issue is not large enough to set out a chapter so we divide each issue
into section. The thesis is empirical study. Hence, we apply models in literature
review which are tested in other countries and this is one of pioneer the studies
applying these tests in Vietnam. This research includes six parts. Part 1 is
introduction of the research with flowing contents: rationales of research, problem
statements, research objectives and research questions, scope and methodologies.
Literature review is mentioned in Part 2. Part 3 describes the detail of
methodologies applying in the research. Data collection and data description are
presented in Part 4. Part 5 is empirical result and Part 6 points out the research
conclusions and implications. The relationship between parts of research is
illustrated in figure 1. 1: Figure 1.1: Structure of research
I. Introduction

II. Literature review

III. Methodology

IV. Data and data description

V. Empirical result

VI. Conclusion

Page 8


II. LITERATURE REVIEW
A great numbers of previous related studies researched on the relationship between
stock markets in the world. We can summary the content of these studies as follow:
studied the interrelationship/linkage amongst the stock markets, examine the long
run cointegration between some stock markets, investigate the benefit/risk of
international diversification and find the leading market amongst developed
countries.
The first topic is studied the interrelation/linkages amongst the stock markets.
Becker et al (1990) examine the intertemporal relation between the US and
Japanese stock markets. They find a high correlation between US stocks’ previous
trading day and the Japanese equity market’s current period performance.
Meanwhile the Japanese market has a small effect on the US return in the current
trading period. Later, Cheung (1992) study the causal relationship between
developed markets and the Asian Pacific markets for the year from 1977 to 1988.
They conclude that the Asia Pacific markets seem to be led by the US market apart
from Thailand, Taiwan and Korea. The Japanese market also impacts these markets
but it is not more important than US.
Following in Cheung (1992) footstep, Chan (1992) seek the relationship of stock
prices in major Asian markets and the US market. They analyze over 1,000
observations of daily data and 224 observations of weekly data of Wednesday’s
closing price for each market of Hong Kong, South Korea, Singapore, Taiwan,
Japan and Standard & Poor’s 500 index (United States) from 1983 to 1987. Their
unit root test results suggest that the individual stock markets are weak form
efficient and no evidence of cointegration is found both single countries’ stock price
and stock price of group countries.

Page 9


After that Park (1993) uses a vector autoregression analysis to analysis the
interrelation between the equity markets of Pacific Basin countries and those of the
US, UK and Japan. There are three results from this research. US are most effective
to Australia. UK and Japan have less interrelation to Pacific Baisan than US. A
group of Hong Kong, Singapore and New Zealand would have more linkages. Bala
Arshanapalli (1995) researches the pre and post-October 1987 stock market
linkages between US and Asian markets. This paper finds the evidences that the
linkages increase since October 1987. It is again minded that the US market are
most influential to Asian equity market compared to that of Japan market. This
integration would be grader during the post October 1987 period.
One more researching related to crisis in year 1987 is Palac-McMiken (1997). They
discover the cointegration amongst ASEAN stock markets. They analyze the data of
monthly ASEAN market index including Singapore, Thailand, Indonesia, Malaysia
and Philippines from 1987 to 1995. Their outcomes are that ASEAN markets are
not efficient for the period studied. However those markets appear to be linked to
each other except for Indonesia market. Janakiramanan (1998) carries out the
linkages between Pacific-Basin stock markets during period 1988 to 1996 by using
a vector autoregression model. They report that US market influence those markets
except for Indonesia.
Later Abul (1999) starts with fluctuations of Asian stock market due mainly to
intra-regional contagion effects based on Asian emerging stock markets and using
vector error-correction model (Toda and Phillips, 1993) and level VAR model
containing integrated and cointegrated processes of arbitrary orders (Toda and
Yamamoto, 1995). They also find a high level of interdependence amongst these
markets in Malaysia, Thailand, Japan, Hong Kong, Singapore and US during the
period from 1992 to 1997. Sheng (2000) surveys cointegration and variance
decomposition amongst national equity indices before and during the period of the

Page 10


Asian financial crisis. They record the evidences of the interrelation between US
and

Asian

markets

during

the

period

of

crisis.

The second content is examine the long run cointegration between some stock
markets such as Andy et al (1995) studies relationship of nine stock markets. The
Engle and Granger cointegration analysis and Granger causality tests are applied to
monthly time series of nine major stock market indices over the period January
1982 to February 1991 to examine for causal linkages. The empirical results
indicate that there are adequate evidences to refute the notion of informationally
efficient stock markets. So no long run relationship exists. Stulz (1996) investigates
the properties of cross country stock return comoverments of US and Japan from
1988 to 1992. They construct overnight and intraday returns for a portfolio of
Japanese stocks using their NYSE-traded American Depository Receipts (ADRs)
and a matched-sample portfolio of US stocks. We find that US macroeconomic
announcements, shocks to the Yen/Dollar foreign exchange rate and Treasury bill
returns and industry effects have no measurable influence on US and Japanese
return correlations. However, large shocks to broad-based market indices (Nikkei
Stock Average and Standard and Poor's 500 Stock Index) positively impact both the
magnitude and persistence of the return correlations. The results suggest that
covariance change over time and can be forecasted using various instrumental
variables. They also conclude that it is not suitable to assume that covariance
between countries are constant.
Ng.(2002) examines the linkages between the South–East Asian stock markets
following the opening of the stock markets in the 1990s. No evidence is found to
indicate a long–run relationship amongst the South–East Asian stock markets over
the period 1988–1997. However, correlation analyses indicated that the South–East
Asian stock markets were becoming more integrated. The results from the time–
varying parameter model also show that the stock market returns of Indonesia, the

Page 11


Philippines and Thailand have all become more closely linked with that of
Singapore.
Jian Yang (2003) examines whether long-run integration between the United States
and many international stock markets has strengthened over time, with special
attention paid to the impact of the abolition of capital control in these markets and
the 1987 international stock market crash with data from (January 1970-December
2001). The results show that no long-run relationship between most of these
markets and the United States exist. However, there are evidences of recent
increasing integration between many smaller markets and the United States while
no such pattern emerges for larger markets including Japan, the United Kingdom,
and Germany, which suggests long-run benefits to U.S. investors of diversifying
into these larger markets.
Ng. (2010) studies an analysis on the long run relationship and risk diversification
amongst Malaysian and Tiger markets (Hong Kong, South Korea, Singapore and
Taiwan) with adopting the Johansen multivariate cointegration test and VECM
using a five variable model and followed Granger causality test. Their findings
conclude that existed the long run relationship amongst the regional stock markets
though such relationships appear to be weak in the short run. The Hong Kong,
South Korea and Taiwan markets influence the Malaysian and Tiger markets and
the Malaysian market affected the Singapore market.
The third content is investigation of benefit/risk of international diversification.
Markridakis (1974) investigates an analysis the interrelationships amongst the
major world stock exchanges. Their purposes are to analyze the potential gains from
international portfolio diversification amongst stock markets of France, Germany,
England, Canada, Australia, Japan, Belgium, Netherlands, Italy, Sweden, and
Switzerland. They also divide the studied period into three sub-periods of bull
market (5/1/1968-29/11/1968), bear market (29/11/1968-26/5/1970) and bull market

Page 12


(26/5/1970-30/9/1970) in three cases of one day lag, no lag-no lead and one day
lead by using the factor analysis and principal components analysis. Two
conclusions drawn from this study are the unstable relationships amongst these
markets and no way to predict the form of these possible interrelationships before
the fact.
Angelos (1999) discovers the long run benefits from international equity
diversification for a UK investor diversifying in the US equity market by using the
MSCI indices from 03/01/1983 to 29/11/1996 (entire period), 03/01/1983 to
30/09/1987 (pre crash period) and 1/11/1987 to 29/11/1996 (post crash period).
Their methodologies are the ADF test with a deterministic trend and cointegration
tests based on the Johansen method. The conclusions are those the long run
diversification benefits for a UK investor diversifying in the US equity market are
reduced during the post crash period. And the long run relationship during the post
crash period exists and no long run relationship during the pre crash period.
David Ely (2001) studies American depositary receipts-analysis of international
stock price movements. They employ the cointegration techniques and estimate
error correction (EC) models to examine the degree of integration between US
(American depositary receipts-ADRs) and three foreign equity markets (UK, Japan
and Germany). They use data from 02/1/1996 to 31/3/1999. They find that ADRs
were cointegrated with ordinary shares trading in three markets. For long term
investors, ADRs are a substitute for ordinary shares trading in foreign markets.
Tzeng (2009) examines the international equity diversification between the US and
its major trading partners (Canada, Germany, Japan and Mexico) with more
powerful nonparametric cointegration test developed by Bierens (1997) from
02/01/2000 to 31/08/2008. The evidences show that the US stock markets are not
pairwise cointegrated with all the stock markets of its major trading partners with
the exception of Mexican stock markets.
Page 13


The fourth content is finding the leading market amongst developed countries. Shim
(1989) studies the international transmission of stock market movements by
estimating a nine market vector autoregression (VAR) system (Australia, Canada,
France, Germany, Hong Kong, Japan, Switzerland, the UK and US) from the period
of 1979 to 1985. Innovations in the US are rapidly transmitted to other markets in a
clearly recognizable fashion, whereas no single foreign market can significantly
explain the US market movements. The evidences imply that US market is a leading
worldwide trend.

Gerrits (1999) studies the short and long term links amongst European (Germany,
UK and Netherlands) and US stock markets for period from 1990 to 1994. Results
of the tests show that the US exerts a significant impact on European markets and
are a leading stock market of this group. Moreover, the three European markets
influence each other in the short and long run. Therefore, diversification amongst
these national stock markets will not greatly reduce the portfolio risk without
sacrificing the expected return.

David (2003) investigates the structure of interdependence in international stock
markets (Australia, Canada, France, Germany, Hong Kong, Japan, Switzerland, the
UK and US) using the error correction model (EC) and directed acyclic graphs
(DAG). The results are the US market is highly influenced by its own historical
innovations and also influenced by market innovations from the UK, Switzerland,
Hong Kong, France and Germany. The US market is the only market that has a
consistently strong impact on price movements in other major stock markets in the
long run.

Page 14


Above group of four contents may not mention about the relationship amongststst
Vietnam stock market and US and related Asian countries. While Vietnam has
joined WTO over four years; many economic operations have changed including
the stock market exchange. Many foreign investors join Vietnam stock market and
number of investors of Vietnam stock market increase strongly. Not so many
researchers study those interrelationships before, during global financial crisis of
2007-2010. This thesis hopes to fill those gaps.

Page 15


III. METHODOLOGY
This part of thesis presents the analytical framework which is applied to find the
answers of the thesis objective. The proxies and data used in this thesis are also
mentioned in this content. With the purpose of investigating the relationship
amongst the stock markets in sample, we follow the other empirical researches in
choosing the methodologies (Chan (1992); Subramanian (2008); Lim (2008); Ng
(Ng 2010); V.X.Vinh (2009)). The Johansen cointegration test is used to analyse the
long run comovement, the Granger causality to test the causality relationship, the
test for converging trend to test for converging or diverging trend of a market index
series, the variance decomposition analysis to describe the causality outside of the
sample estimation.

1. Johansen Cointegration test
In statistics, the Johansen test is a procedure for testing cointegration of several I(1)
time series. As the pioneering work by Engle and Granger (1987) reports that if the
individual economic series are stationary only after differencing but a linear
combination of their levels is stationary and the series are said to be cointegrated.
This test permits more than one cointegrating relationship so it is more generally
applicable than the Engle–Granger test which is based on the Dickey–Fuller (or the
augmented) test for unit roots in the residuals from a single (estimated)
cointegrating relationship. Johansen (1988) and Johansen and Juselius (1990)
propose a maximum likeihood method for estimating long run equilibrium
relationships or cointegrating vectors and derive likelihood ratio tests for
cointegration. There are two types of Johansen test: trace test or eigenvalue test. The
null hypothesis for the trace test is the number of cointegration vectors r ≤ n, the
null hypothesis for the eigenvalue test is r = n.

Page 16


The purpose of applying the cointegration test in this thesis is to find the
cointegration (long run relationship) amongst stock markets in the studied sample.
Before conducting the cointegration test, unit root test need to be analyzed the
stationary of time series variables.
1.1 Unit root test
An econometric model of cointegration requires knowledge of stationary and order
of integration for the time series variables. Condition for cointegration that all the
time series to be analyzed is integrated of the same order or that all series contain a
deterministic trend (Granger, 1986). Thus, prior to conducting the cointegration
analysis of stock markets, it is necessary to examine the order of integration of each
stock index. There are some methods of testing for stationary such as visual plots of
data; the autocorrelation function or unit root test. The unit root test is the most
widely used test for stationary. For this reason, unit root test is applied in this
studying. Three common tests which employed to test for unit root in this research
are Dickey-Fuller test (Dickey and Fuller, 1979; Fuller 1976), Augmented Dickey
Fuller test (1979, 1981) and Phillips Perron test (1988). These three unit root tests
are employed by several previous researches (V.X.Vinh (2009); David (2001)). The
presence of a unit root reveals that the time series data is non stationary. In order to
make time series data stationary, it is differenced d times or integrated of order d
(I(d)). Cointegration procedure requires the variables are stationary at the first
difference I(1).
1.1.1 The Dickey-Fuller test (DF)
To apply the DF test, let consider a simple autoregression:
yt = ρ yt −1 + δ xt + ε t (1)

Where y is a non stationary time series. ρ and δ are parameters to be estimated. xt
are optional exogenous regressors that may include a constant or a constant and

Page 17


trend. ε t are sequence of independent normal random variables with mean zero and
a constant variance. The time series yt is a non stationary and has unit root if the
absolute value of ρ equals one. If the absolute value of ρ less than one, the time
series yt is a stationary. Hence, the basis objective of unit root test is to test the null
hypothesis that ρ =1. So we have:
Null hypothesis H 0 : ρ =1
Alternative hypothesis H1 : ρ <1
To apply the standard DF test, we subtract yt −1 from both sides of the equation (1).
Equation (1) becomes:
∆yt = α yt −1 + xt'δ + ε t (2)

Where α = ρ -1
So null hypothesis and alternative hypothesis are:
Null hypothesis H 0 : α =0 ( ρ =1)
Alternative hypothesis H1 : α <0 ( ρ <1)
Above hypothesises can be evaluated using the conventional t-ratio.
tα =

αˆ
SE (αˆ )

Where tα is a t-ratio for α , αˆ is the estimate of α and SE (αˆ ) is the coefficient
standard error.
1.1.2 The Augmented Dickey-Fuller test (ADF)
The DF test are only valid if ε t are white noise. If the dependent variables of the
equation (2) which we do not model are autocorrelation, ε t will be autocorrelated.
Said and Dickey (1984) augment the test using p lags of the dependent variable y to
the right hand of the equation (2):

Page 18


Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Tải bản đầy đủ ngay

×