INTRODUCTION
Research background
The stock market is a crucial element of any economy's financial system, significantly impacting both developed and developing nations It plays a vital role in mobilizing savings, effectively connecting savers with investors By providing essential services, the stock market benefits governments, businesses, and investors alike Governments can issue bonds to fund infrastructure projects, while corporations can raise capital through initial public offerings (IPOs) Additionally, the stock market allows investors to diversify their portfolios, thereby mitigating risks associated with investing in various stocks Consequently, stock prices are a matter of widespread concern for governments, enterprises, and investors, leading to a keen interest in understanding the factors that influence these prices globally.
In economic theory, stock prices are primarily influenced by the relationship between supply and demand When demand for a share exceeds its supply, the share price rises; conversely, if supply surpasses demand, the price declines Beyond these microeconomic factors, numerous studies have explored the impact of macroeconomic variables—such as industrial production, interest rates, inflation, and exchange rates—on stock prices Research conducted in developed markets like the U.S and Japan has demonstrated that these macroeconomic factors significantly affect stock prices This body of work has encouraged further investigation into similar dynamics in emerging markets, including Singapore, Thailand, Malaysia, Indonesia, the Philippines, and Vietnam.
Ngoc, 2009), and Jordan (El-Nader & Alraimony, 2012)
The Vietnamese stock market, consisting of the Ho Chi Minh Stock Exchange (HOSE) and the Hanoi Stock Exchange (HNX), is recognized as one of Asia's emerging markets, with both exchanges being established in July 2000 and March 2000, respectively.
2005, respectively and they have made significant contributions to the economy
The Vietnamese stock market, like other emerging Asian markets, is gaining attention for its potential to yield significant returns for investors Following a period of rapid growth in stock prices throughout 2007, the market experienced a notable decline after that year, which was characterized by a stock market bubble that peaked at 1170.7 points in March 2007 Various factors contribute to the fluctuations in stock prices, including both international and domestic economic uncertainties and investor psychology Despite the market's potential, there is a lack of empirical research examining the relationship between stock prices and macroeconomic indicators, raising questions about the impact of these indicators on Vietnamese stock prices.
Figure 1.1 VN-Index from January 2001 to May 2013
The Ho Chi Minh Stock Exchange (HOSE) provides the latest updates and resources for downloading comprehensive thesis documents For further inquiries or access to full materials, please contact via email at vbhtj mk gmail.com.
Research problems
Numerous studies have explored the influence of macroeconomic variables on stock prices, demonstrating their significant impact Notable research includes Chen et al (1986) focusing on the United States and Mukherjee and Naka (1995) examining Japan.
Wongbangpo and Sharma (2002) conducted research on five Asian countries—Singapore, Thailand, Malaysia, Indonesia, and the Philippines—while also examining Turkey in 2005, and El-Nader and Alraimony (2012) focused on Jordan The findings from these studies reveal significant discrepancies across different nations Specifically, Wongbangpo and Sharma (2002) identified a negative relationship between interest rates and stock prices in Singapore, Thailand, and the Philippines, whereas a positive relationship was observed in Indonesia and Malaysia.
There is a limited amount of research on how macroeconomic indicators affect stock prices in emerging markets, particularly in Vietnam (Hussainey & Ngoc, 2009) Hussainey and Ngoc's study is the first to explore the relationship between macroeconomic indicators and Vietnamese stock prices, making a significant contribution to the existing literature However, their research is limited to only two domestic macroeconomic indicators—interest rates and industrial production—along with US macroeconomic factors They recommend including additional macroeconomic variables in future studies to enhance the validity of the results.
Research objectives
This study aims to analyze the influence of four critical macroeconomic indicators—industrial production index, interest rates, inflation rates, and exchange rates—on stock prices in Vietnam Specifically, it seeks to address key questions regarding their impact on the Vietnamese stock market.
● Do the long run and short run relationship between selected macroeconomic indicators and Vietnamese stock prices exist?
● How do macroeconomic indicators affect the stock price in long run and short run?
Significance of the research
The Vietnamese stock market, being relatively immature, requires extensive research to enhance understanding and promote sustainable development This paper aims to explore the relationship between macroeconomic indicators and stock prices, providing valuable insights for policymakers, company managers, and investors The findings are expected to assist in predicting stock price fluctuations in response to changes in macroeconomic conditions.
This study enables policymakers to assess the effects of fiscal and monetary policies on the stock market, allowing for the development of more effective strategies to timely and efficiently regulate the Vietnamese stock market.
This study not only aids policymakers but also provides valuable insights for managers of publicly listed companies, enabling them to comprehend how external factors influence their stock prices and maintain stability in those prices.
Moreover, the findings drawn from this study can help the investors understand more about the volatility of the stock market as a measure of risk
Investors can enhance their decision-making and minimize risks by leveraging macroeconomic insights For example, if research indicates that rising inflation negatively impacts stock prices, it is advisable for investors to rebalance their portfolios by selling stocks and reallocating funds to more lucrative assets.
Research methodology and scope
This study examines the relationship between stock prices on the Ho Chi Minh Stock Exchange (VN-Index) and four key macroeconomic variables: industrial production, interest rate, inflation rate, and exchange rate The analysis does not include the HNX-Index or its relationship with these variables Utilizing 65 monthly observations from January 2008 to May 2013, the study follows a structured approach for data analysis.
● Using ADF test to test the stationary of all time series data
The cointegration test is employed to determine the existence of a long-run relationship among variables.
● Applying Vector Error Correction Model to investigate the long run as well as the short run relationship if there is a cointegration relationship among variables
Eviews software version 6.0, SPSS version 16.0 software and Microsoft Excel software are used as data analysis tools.
Research structure
This research is divided into different chapters and each chapter covers some areas of the research The structure of the research is as follows:
Chapter 1: Introduction: This chapter depicts general information about the research including research background, research problems, research objectives, significance and limitations of the research
Chapter 2: Literature review: The relevant literature on the research is reviewed in this chapter, including financial theories and previous empirical studies regarding the relationship between macroeconomic factors and stock prices In addition, the hypotheses, which are tested in this research, are also set
Chapter 3: Research methodology: This chapter discusses about the research process, measurement of variables, data collection and methods of data analysis employed in this study
Chapter 4: Data analysis and results: Collected data are analyzed in this part in order to investigate the relationship between macroeconomic variables tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg and stock price In addition, the results obtained from empirical research are also analyzed
Chapter 5: Conclusions and implications: This chapter is all about conclusions of this study, implications, limitations, and recommendations for further studies relating to the research topic. tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
LITERATURE REVIEW
Theoretical framework
This study explores the connection between macroeconomic variables and stock prices through two key theories: the top-down approach and the dividend valuation model, also known as the present value model.
The top-down approach in the valuation process is essential for determining the intrinsic value of ordinary shares, as it involves gathering and organizing information about the economy, industry, and company (Gitman et al., 2004) This method consists of three key steps: economic analysis, industry analysis, and company analysis It emphasizes the significant influence of macroeconomic variables on both the company and its stock price, highlighting the relationship between these factors and stock valuations.
The intrinsic value of an investment, as defined by Gitman et al (2004), is determined by the present value of anticipated cash flows For a stock, this intrinsic value comprises the annual cash dividends received and the expected future price of the stock.
The cash flow benefits of a share can be understood by considering the dividends received over an infinite period In this view, a share's value is determined by the present value of all future dividends expected indefinitely This method, which posits that a share's worth is based on its future dividends, is referred to as the dividend valuation model (DVM).
The DVM can be expressed as the following equation:
P is the value of the share
D t is the dividend received at year t r is the required rate of return
The value of a share is determined by the present value of all future dividends or expected cash benefits, as indicated by the equation Consequently, any economic factor that impacts the expected future cash flow or the required rate of return will also affect the share's value.
Relationship between industrial production and stock price
Industrial production serves as a key indicator of real economic activity, reflecting the pace of economic growth or contraction compared to previous years Typically, industrial production rises during periods of economic expansion and falls during recessions, signaling shifts in the economy As the productive capacity of an economy increases with growth, corporations can generate higher cash flows, which directly influence stock prices Significant changes in economic growth impact investor decisions; for example, if investors perceive economic growth and increased corporate profits, they are likely to invest more in stocks, driving prices up Conversely, a decline in economic growth may lead investors to sell stocks or offer lower prices, resulting in decreased stock values.
Wongbangpo and Sharma (2002) assert that an increase in output leads to higher expected future cash inflows and profitability for companies, resulting in increased future dividends This expectation encourages investors to purchase shares at elevated prices, driving up share prices Conversely, during a recession, a decline in output negatively impacts profitability, causing share prices to fall In summary, the theory indicates a positive relationship between stock prices and industrial production.
Many empirical studies utilize the industrial production index as a proxy for economic activity, suggesting that production growth aligns with average company sales and cash flows This index is deemed valuable in asset pricing models (Chen et al., 1986) Additionally, research in emerging markets indicates a positive correlation between real economic activity and stock prices Recent studies, such as those by El-Nader and Alraimony (2012), investigate the relationship between the Amman stock market returns and Jordanian macroeconomic factors, highlighting that real gross domestic product (GDP) positively influences stock prices in Jordan.
A study by Wongbangpo and Sharma (2002) across five Asian countries—Indonesia, Malaysia, the Philippines, Singapore, and Thailand—reveals a positive correlation between stock prices and economic growth, as indicated by gross national product (GNP).
A study conducted by (2005) examines the relationship between industrial production and stock prices on the Istanbul Stock Exchange in Turkey from 1991 to 2000 The findings indicate that industrial production positively influences stock returns, except during the period from the onset of the 1994 financial crisis to the beginning of the 1997 Asian crisis.
In the case of Vietnam, Hussiney and Ngoc (2009) examine the relationship between industrial production and stock prices in Vietnam from 2001 to 2008
They find out that industrial production has a positive effect on Vietnamese stock prices Their finding is consistent with theoretical expectations as well the results of previous studies
Based upon the theories and empirical findings drawn from previous studies, the following hypothesis is set:
H1 Industrial production has a positive impact on the stock price in Vietnam.
Relationship between interest rate and stock price
In stock valuation, investors first establish a discount rate, which reflects their required rate of return This discount rate is influenced by two key factors: the time value of money and the stock's risk level The time value of money is typically represented by the risk-free rate.
Figure 2.1 VN-Index and Industrial Production Index
The risk premium serves as compensation for the inherent risk associated with stocks A widely used approach to calculate the required rate of return is the Capital Asset Pricing Model (CAPM).
R i : the required rate of return on investment i
R f : the risk-free rate of return, the return that can be earned on a risk-free investment β i : beta coefficient, or index of non-diversifiable risk, for investment i
R m : the market return; the average return on all securities
Changes in interest rates serve as a proxy for fluctuations in the risk-free rate, which subsequently impacts the discount rate According to Mukherjee & Naka (1995), an increase in the risk-free rate results in a higher required rate of return, as illustrated in equation 2.2 This increase leads to a decline in share prices, based on the dividend valuation model (DVM) Conversely, if interest rates decrease while all other factors remain constant, share prices are likely to rise due to a lower required rate of return.
Interest rates significantly influence investor behavior in the stock market, particularly regarding asset portfolio allocation An increase in interest rates raises the opportunity cost, prompting investors to shift from equities to other assets Apergis and Eleftheriou (2002) highlight that when investors hold both bonds and stocks, higher interest rates lead them to favor bonds over stocks, resulting in a tendency to buy bonds and sell stocks.
As a result, the stock prices will decline In contrast, a decrease in interest rate leads to an increase in stock prices
In addition, the interest rate is also considered as the cost of capital, which is the price paid for the utilization of money for a specific period of time
Rising interest rates negatively impact corporate profitability, particularly for companies that rely on borrowed funds for operations Increased interest expenses can lead to lower profits and reduced future cash inflows, causing stock prices to decline If interest expenses escalate to a level that threatens insolvency, the risk of bankruptcy increases, prompting investors to demand higher risk premiums, which further depresses share prices Additionally, higher costs of capital affect investors who borrow to invest in stocks, reinforcing the negative correlation between stock prices and interest rates.
Numerous empirical studies have explored the relationship between interest rates and stock prices in both developed and emerging markets In Bangladesh, Uddin and Alam (2007) identified a significant negative correlation between interest rates and stock prices Similarly, a recent study covering South Asian nations, including Pakistan, India, and Sri Lanka, found that interest rates negatively and significantly affect the stock markets in these countries (Aurangzeb, 2012).
Wongbangpo and Sharma (2002) conducted a study on several Asian countries, revealing mixed results about the relationship between interest rates and stock prices They found a negative correlation in Singapore, the Philippines, and Thailand, while Malaysia and Indonesia exhibited a positive relationship.
Hussianey and Ngoc (2009) conducted a study in Vietnam using monthly data from 2001 to 2008 to examine the relationship between interest rates and stock prices They used basic interest rates as a proxy for short-term rates and ten-year government bond rates for long-term rates Their findings revealed that long-term and short-term interest rates do not influence Vietnamese stock prices in the same direction.
The research indicates that short-term interest rates are positively correlated with stock prices, whereas long-term interest rates negatively affect stock prices.
Theories indicate a negative relationship between stock prices and interest rates, yet empirical evidence remains inconclusive Hussianey & Ngoc (2009) discovered a positive impact of short-term interest rates on stock prices in Vietnam from 2001 to 2008 This study anticipates that this relationship will persist in subsequent periods.
Additionally, as the figure 2.2 shows, the stock prices and interest rates seem to move in the same direction from 2008 to 2013 Therefore, this study proposes the the following hypothesis:
H2 Interest rate has a positive impact on the stock price in Vietnam.
Relationship between inflation and stock price
Inflation is another factor commonly used in previous empirical studies to investigate the relationship between macroeconomic indicators and stock prices
According to the Fisher effect, the real interest rate is equal to the nominal interest
Figure 2.2 VN-Index and Interest Rate
Inflation significantly influences the real return on investments by reducing the expected returns when it rises It also affects the required rate of return for investors, as it alters the nominal interest rate Overall, inflation impacts stock prices in two primary ways: first, by affecting future real earnings, and second, by influencing how investors discount those future earnings.
Inflation significantly influences investment decisions and economic growth, as rising inflation can lead to a decline in real income, prompting investors to sell financial assets and causing stock prices to drop Conversely, low inflation encourages investors to purchase more financial assets, including stocks The relationship between inflation and stock prices is primarily driven by its effect on a company's earnings; with low inflation, companies can maintain lower costs and increase profits, making investors more inclined to buy stocks at higher prices.
Inflation impacts stock prices primarily through the discount factor, which is composed of the risk-free rate and the risk premium, as outlined by the Capital Asset Pricing Model (CAPM) When inflation rises, interest rates increase, leading to a higher risk-free rate This elevation in the discount rate results in a decreased present value of future earnings, consequently causing stock prices to decline Therefore, the theory indicates a negative correlation between inflation rates and stock prices.
Udegbunam and Eriki (2001) investigate the impact of inflation on stock prices in Nigeria, utilizing a straightforward stock price model that incorporates key determinants as control variables, estimated through the Ordinary Least Squares method Their research reveals that inflation adversely affects the Nigerian stock market through multiple channels, including discount rates, nominal contracts, and tax effects.
A recent study by El-Nader and Alraimony (2012) reveals a negative relationship between the Amman stock index and inflation in Jordan This finding aligns with similar results observed in five Asian countries—Thailand, Indonesia, Malaysia, Singapore, and the Philippines (Wongbangpo & Sharma, 2002)—as well as in Ghana (Coleman & Tettey, 2008).
Aurangzeb (2012) analyzes data from 1997 to 2003 across Pakistan, India, and Sri Lanka to explore the relationship between macroeconomic indicators, particularly inflation, and stock prices The study reveals an insignificant negative impact of inflation on the stock market performance in these countries Despite empirical findings and theories suggesting a negative relationship between inflation and stock prices, this research proposes a hypothesis based on previous studies and theoretical frameworks.
H3 Inflation rate has a negative impact on the stock prices in Vietnam.
Relationship between exchange rate and stock price
Previous studies highlight the significance of exchange rates in analyzing the influence of macroeconomic indicators on stock prices, as they directly affect a firm's cash flow Specifically, fluctuations in exchange rates impact exporting and importing companies differently When the local currency depreciates, exported goods become more appealing due to lower prices, potentially increasing demand If the demand for exports and imports is elastic, this can lead to a rise in the volume of exports, resulting in enhanced cash flows, profits, and stock prices for exporting firms.
Figure 2.3 VN-Index and CPI
The fluctuation of exchange rates significantly impacts imported products, leading to increased costs that reduce cash flows, profits, and ultimately the stock prices of importing firms As a result, the stock prices of these companies tend to decline Thus, the relationship between exchange rates and stock prices can be either positive or negative, depending on various economic factors.
The rise of economic globalization has significantly impacted businesses, making them susceptible to international activities Consequently, fluctuations in exchange rates affect both multinational and domestic firms, with multinational companies experiencing more immediate consequences A change in exchange rates can lead to profits or losses in foreign operations if not hedged, ultimately altering the firm's value and influencing its stock price.
Foreign investors in the stock market consider both stock returns and the stability of the host currency, as exchange rate fluctuations significantly impact their investment decisions A depreciation of the host currency can reduce the value of shares held by foreign investors when converted back to their home currency If stock returns do not offset the losses from currency depreciation, investors risk losing money, prompting them to sell their stocks to protect their capital This selling pressure can lead to a decline in stock prices.
El-Nader and Alraimony (2012) examine the connection between exchange rates and stock prices in Jordan, highlighting that fluctuations in exchange rates negatively affect the Amman stock market Their findings align with those of Aurangzeb, reinforcing the significance of understanding this relationship for investors and policymakers.
A study by Aurangzeb (2012) reveals a positive relationship between exchange rates and stock prices in South Asian countries, specifically Pakistan, India, and Sri Lanka.
Wongbangpo and Sharma (2002) investigate the varying impacts of exchange rates on stock prices across Southeast Asia, revealing a positive relationship in Indonesia, Malaysia, and the Philippines, while identifying a negative correlation in Singapore and Thailand.
A study by & (2005) on the Turkish stock exchange reveals that the depreciation of the domestic currency resulted in decreased stock returns prior to and during the 1994 financial crisis; however, this relationship reversed in the aftermath Overall, the empirical evidence regarding the connection between exchange rates and stock prices is inconsistent, indicating that the relationship can be either positive or negative.
Vietnam's status as an importing country means that a depreciation of its domestic currency leads to higher costs for imported goods This increase in expenses negatively affects cash flows, profits, and ultimately the stock prices of firms reliant on imports As a result, the study posits that the exchange rate exerts a negative influence on Vietnamese stock prices, leading to the following hypothesis.
The exchange rate negatively affects stock prices in Vietnam.
Hypotheses summary
After reviewing the theories as well as previous empirical studies, the hypotheses regarding the relationship between stock prices and macroeconomic factors are summarized as follows:
H1 Industrial production has a positive impact on the stock prices in Vietnam
H2 Interest rate has a positive impact on the stock prices in Vietnam
H3 Inflation rate has a negative impact on the stock prices in Vietnam
H4 Exchange rate has a negative impact on the stock prices in Vietnam
Figure 2.4 VN-Index and Exchange Rate
Source: HOSE and IFS tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Research model
This study develops a research model informed by prior studies and hypotheses, incorporating the frameworks established by Coleman and Tettey (2008) and Hussiney and Ngoc (2009).
H2 (+) H3 (-) H4 (-) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
RESEARCH METHODOLOGY
Research process
Research process depicts steps need to be conducted in this study The research process in this study was designed as follows:
Literature review Relevant theories Previous empirical studies
Research design tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Measurement of variables
Vietnamese stock prices are primarily represented by the VN-Index for the Ho Chi Minh Stock Exchange (HOSE) and the HNX-Index for the Hanoi Stock Exchange (HNX) This study focuses on the monthly VN-Index as a proxy for Vietnamese stock prices and as a dependent variable for empirical analysis, selected based on factors such as capitalization scale, operating time, the number of listed companies, and the number of investors The monthly VN-Index is calculated using the natural logarithm of the monthly closing price (Coleman & Tettey, 2008; Pal & Mittal).
In this empirical study, industrial production, interest rates, inflation rates, and exchange rates were utilized as independent variable proxies Industrial production was quantified using the month-end industrial production index (Hussiney & Ngoc, 2009) The interest rate was represented by the monthly lending rate (Coleman & Tettey, 2008) Inflation rates were assessed through the month-end consumer price index compared to the same period in the previous year (El-Nader & Alraimony, 2012) Lastly, the exchange rate was determined by the monthly average exchange rate, reflecting the amount of domestic currency (VND) per unit of USD (El-Nader).
All independent variables were transformed using the natural logarithm, as noted by Coleman and Tettey (2008) and Pal and Mittal (2011).
Data collection and sample size
Data collection methods can be categorized into primary and secondary data collection Primary data is gathered from firsthand experiences and is considered more reliable since it has not been previously published In contrast, secondary data consists of information that has already been published or utilized in various forms, such as literature reviews in research studies, journals, articles, and publicly available internet sources.
This study employed a secondary data method, utilizing a time series of monthly data from January 2008 to May 2013 Vietnamese stock prices (VN-Index) were sourced from the Ho Chi Minh Stock Exchange, while industrial production data was obtained from the General Statistic Office (GSO) of Vietnam Additional variables, including interest rates, CPI, and exchange rates, were collected from the International Financial Statistics database published by the International Monetary Fund.
Table 3.1 Description of variables Variables Definitions Symbols Calculations Sources
Stock price VN-Index LVNI ln(VNI t ) HOSE
Monthly index of industrial production LIP ln(IP t ) GSO
Interest rate Monthly lending interest rate LIR ln(IR t ) IFS-IMF
Inflation rate Monthly consumer price index LCPI ln(CPI t ) IFS-IMF
The article discusses the exchange rate between the Vietnamese Dong (VND) and the US Dollar (USD), highlighting the importance of understanding foreign exchange rates (EXR) in the context of international financial systems It emphasizes the role of institutions like the International Monetary Fund (IMF) in providing insights and data for economic analysis Additionally, it mentions the availability of resources for downloading the latest research and theses related to this topic.
Model specification
This study utilized the empirical model proposed by Coleman and Tettey (2008) to analyze the relationship between the selected variables and the stock price index The following empirical model was estimated for this purpose.
The stock price index (VNI\_t) is influenced by several economic factors, including the industrial production index (IP\_t), lending interest rate (IR\_t), consumer price index (CPI\_t), and exchange rate (EXR\_t) The coefficients of these variables are represented by β\_1, β\_2, β\_3, and β\_4, while ɛ\_t denotes the error term in the model.
This study applied the natural logarithm to all variables in equation 3.1 to facilitate a partial elasticity analysis (Coleman and Tettey, 2008) This approach allows for the assessment of the impact of changes in macroeconomic variables on the stock price index while holding other factors constant Consequently, the estimated equation is presented as follows:
Method of data analysis
This part of the study explained the methods applied to analyze the data for the purpose of this thesis This thesis followed the method introduced by Coleman
In their research, Tettey (2008) applied the unit root test to determine the stationarity of time series data, as non-stationary data can lead to spurious regression results (Gujarati, 2003) If the variables are found to be non-stationary at levels but become stationary after differencing, a cointegration test is necessary to explore the long-run relationships among them Should the cointegration test indicate that the time series variables are cointegrated, the Vector Error Correction Model (VECM) is the most effective method for modeling the relationships between these variables (Coleman & Tettey, 2008).
The unit root test is a widely used method for assessing the stationarity of time series data, which can be classified as either stationary or non-stationary As noted by Gujarati (2003, p.797), a time series is considered stationary if its mean and variance remain constant over time, and the covariance between two time periods is determined solely by the lag or distance between them.
The concept is independent of the specific time when the covariance is calculated Conversely, a process is considered non-stationary if its mean and variance fluctuate over time.
The importance of stationarity in time series analysis cannot be overstated, as non-stationary time series can only be examined within a specific time frame, limiting their generalizability to other periods (Gujarati, 2003) To avoid spurious regression, it is essential to conduct a pretest to confirm a stationary cointegration relationship among variables In this study, all time series data were analyzed for unit roots prior to testing the model and hypotheses The Augmented Dickey Fuller (ADF) test, developed by Dickey and Fuller, was employed to assess the stationary properties of the time series data, as outlined by Gujarati (2003).
Where, α, β, and δ are coefficients ɛ t is the error term m is the number of lags
Hypothesis testing for ADF test is as follows:
The null hypothesis \( H_0: \delta = 0 \) indicates that the time series is non-stationary, implying the presence of a unit root, while the alternative hypothesis \( H_1: \delta < 0 \) suggests that the time series is stationary and does not contain a unit root To determine the presence of a unit root, the t-value of the coefficient of \( Y_{t-1} \) is calculated and compared against critical values at significance levels of 1%, 5%, and 10% If the estimated t-value is statistically significant and less than the critical values, the null hypothesis is rejected, indicating that the time series is stationary Conversely, if the estimated t-value exceeds the critical values, the null hypothesis cannot be rejected, confirming that the time series is non-stationary In cases of non-stationary time series, it is essential to transform the data into a stationary series, typically achieved by taking the first differences of the non-stationary time series.
In time series analysis, a series \( Y_t \) is classified as I(0) or integrated of order 0 if it is stationary at levels without any differencing Conversely, a series that achieves stationarity after first differencing, represented as \( \Delta Y_t = Y_t - Y_{t-1} \), is termed I(1) or integrated of order 1 More broadly, a time series that becomes stationary after \( p \) differencing operations is referred to as I(p) or integrated of order \( p \).
As mentioned above, the regression of one non-stationary time series on another non-stationary time series may produce a spurious regression (Gujarati,
Differenced time series is commonly employed to mitigate the issue of spurious regression, as noted by Coleman and Tettey (2008) However, this approach primarily captures short-run relationships, neglecting long-run responses Engle and Granger's concept of cointegration addresses the behavior of non-stationary variables, indicating that if two variables, Y and X, are both integrated of order one, I(1), a regression of Y on X can be conducted.
The equation (3.4) can be reformulated to enhance clarity and understanding For the latest updates, please download the full thesis document at the provided email address.
If u t , which is the residuals obtained from the regression (3.4), is stationary, a regression of Y on X would be meaningful (i.e., not spurious) In this case, Y and
X are cointegrated, indicating a long-term equilibrium relationship between the variables, as defined by Gujarati (2003) A regression model, such as (3.4), is referred to as a cointegrating regression, with the slope parameter \$\beta_2\$ representing the cointegrating parameter This concept can also be applied to regression models with multiple regressors, resulting in \$k\$ cointegrating parameters.
According to Gujarati (2003), various methods exist for testing cointegration, including the Dickey Fuller (DF) or Augmented Dickey Fuller (ADF) tests on residuals from cointegrating regression, as well as the cointegrating regression Durbin-Watson (CRDW) test This study utilized the Johansen and Juselius (1990) method, which employs two likelihood ratio test statistics—the Trace test and the Maximum Eigenvalue test—to assess the number of cointegration vectors The Maximum Eigenvalue statistic evaluates the null hypothesis of \( r \) cointegrating relations against the alternative hypothesis of \( r+1 \) cointegrating relations for \( r = 0, 1, 2, \ldots, n-1 \).
Where λ is the Maximum Eigenvalue and T is the sample size
The Trace test examines the null hypothesis of the number of cointegrating relations, denoted as \( r \), against the alternative hypothesis of \( n \) cointegrating relations, where \( n \) represents the total number of variables in the system The test evaluates the hypotheses for \( r = 0, 1, 2, \ldots, n-1 \).
Its equation is computed as the following formula:
If the estimated Trace statistic or Maximum Eigenvalue statistic exceeds the critical values and is statistically significant, the null hypothesis of r cointegrating relations is rejected, indicating the presence of r+1 cointegrating relations among the variables.
Gul and Ekinc highlight that the Trace test and Maximum eigenvalue test can produce differing outcomes, recommending that when discrepancies arise, the results of the Trace test should be prioritized (Asari et al., 2011).
After conducting cointegration test, if cointegration has been detected among series, there exists a long-run equilibrium relationship among them
The vector error correction model (VECM) is utilized to assess both short-run and long-run relationships among cointegrated macroeconomic variables, as noted by Pal and Mittal (2011) This study employs the VECM framework established by Johansen and Juselius (1990), beginning with a Vector Autoregression (VAR) of order m.
The article discusses a model where the variables are represented as an \(n \times 1\) vector, and each \(B_i\) is an \(n \times n\) matrix of parameters It also mentions that \(m\) denotes the number of lags, and the model includes a residual term.
In order to use Johansen’s method, equation (3.8) needs to be turned into a VECM format, which can be written as follows:
Where , and I is an (n x n) identical matrix; ; is a symbol of difference operator
DATA ANALYSIS AND RESULTS
Descriptive statistics
LVNI LIP LIR LCPI LEXR
Source: Eviews 6.0 outcomes tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Table 4.1 presents the descriptive statistics for five key variables: stock price, industrial production index, lending interest rate, consumer price index, and exchange rate, all of which exhibited positive means The skewness and kurtosis values suggest a lack of symmetry in their distributions According to Naik & Padhi (2012), a normal distribution is characterized by skewness and kurtosis values of zero and three, respectively The interest rate's positive skewness indicates a rightward skew, while the negative skewness of the stock price, industrial production index, CPI, and exchange rate points to a leftward skew.
The kurtosis values for interest rate, CPI, and exchange rate are below three, indicating flatter distributions at the mean compared to a normal distribution In contrast, stock price and industrial production index exhibit more peaked distributions, with kurtosis values exceeding three Additionally, the p-values from the Jarque-Bera statistics reveal that the distributions of most variables are not normal, with the exception of interest rate and CPI.
The analysis indicates that the interest rate and Consumer Price Index (CPI) exhibited greater volatility compared to stock prices, the industrial production index, and exchange rates.
Correlation analysis
LVNI LIP LIR LCPI LEXR
Notes:*, ** Correlation is significant at the 0.05 and 0.01 level, respectively (2-tailed)
A key assumption of regression models is that independent variables, or regressors, should not be mutually correlated When multiple independent variables exhibit high correlation with one another, this condition is known as multicollinearity To address this issue, it is advisable to include only one of the highly correlated variables in the model This study conducted a correlation analysis to identify potential multicollinearity among the independent variables A commonly accepted guideline is that if the pairwise correlation between two regressors exceeds 0.8, multicollinearity is present, leading to unreliable results.
Table 4.2 indicates that all correlation coefficients among independent variables were below 0.8, suggesting no multicollinearity issues Additionally, the industrial production index and interest rate showed a positive correlation with the VN-Index, whereas the consumer price index and exchange rate negatively affected the VN-Index.
Unit root test
Critical value at 1% significance level
Table 4.4 Result of unit root test after first differencing
Critical value at 1% significance level
Tables 4.3 and 4.4 summarize the unit root test results for all independent and dependent variables using the Augmented Dickey Fuller (ADF) test, conducted without a constant or linear trend The analysis revealed that all variables were non-stationary at levels, as the estimated statistic values exceeded the critical values at the 1 percent significance level However, after taking first differences, the variables became stationary, with all estimated statistic values falling below the critical values This indicates that all variables are individually integrated of the same order, I(1) Consequently, a cointegration test is necessary to determine whether these variables are cointegrated.
Cointegration test
The unit root tests revealed that all variables were integrated of the same order, specifically I(1) Consequently, the study proceeded to analyze the cointegration among these variables to explore the long-run equilibrium relationship between them.
Before conducting cointegration tests, it is essential to determine the optimal lag length for all variables This study employed a vector autoregressive (VAR) lag order process, utilizing the Schwarz information criterion (SC) and the Hannan-Quinn information criterion (HQ) to identify the optimal lag length for the cointegration and VECM tests The optimal lag length is indicated by the lowest values of SC and HQ.
Table 4.5 VAR lag order selection criteria
Lag LogL LR FPE AIC SC HQ
Notes: * indicates lag order selected by the criterion LR: sequential modified LR test statistic (each test at 5% level) FPE: Final prediction error
AIC: Akaike information criterion SC: Schwarz information criterion HQ: Hannan-Quinn information criterion Source: Eviews 6.0 outcomes
Table 4.5 indicates that the LR and FPE criteria selected four lags, while the AIC selected five lags, and both the SC and HQ criteria chose one lag based on their lowest values Consequently, this study utilized one lag for testing cointegration and for the Vector Error Correction Model (VECM) based on the SC and HQ criteria.
After determining the optimal lag length, the presence and quantity of cointegrating relationships among the selected variables were analyzed using a vector error correction model (VECM) based on Johansen’s method The analysis employed the trace statistic and maximum eigenvalue statistic to identify the number of cointegrating vectors.
Table 4.6 Result of cointegration test Unrestricted Cointegration Rank Test (Trace)
No of CE(s) Eigenvalue Statistic Critical Value Prob.**
Notes: Trace test indicates 1 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegration Rank Test (Maximum Eigenvalue)
No of CE(s) Eigenvalue Statistic Critical Value Prob.**
Notes: Max-eigenvalue test indicates no cointegration at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values Source: Eviews 6.0 outcomes
The Trace test reveals the presence of one cointegrating equation, while the Maximum Eigenvalue test shows no evidence of cointegration among the variables at the 0.05 significance level In cases where the Trace and Maximum Eigenvalue tests yield differing results, the Trace test is generally preferred (Gul and Ekinc, as cited in Asari et al.).
In 2011, this study utilized the Trace test, which revealed that the null hypothesis of no cointegrating relation was rejected at the 0.05 significance level According to Table 4.6, there is one cointegrating relation among the variables, indicating a long-run equilibrium relationship between stock prices and macroeconomic variables.
Hypotheses testing
The unit root and cointegration tests reveal that all variables are integrated of the same order, indicating a long-run equilibrium relationship among them Consequently, the Vector Error Correction Model (VECM) was employed to analyze both the long-run and short-run relationships among the selected variables According to equation (3.10), the VECM effectively illustrates the interactions between stock prices and macroeconomic variables.
The article discusses key components of a vector error correction model, including a constant, the coefficient of vector error correction, long-run coefficients, short-run coefficients, and residuals.
Table 4.7 Result of cointegrating vector
LVNI LIP LIR LCPI LEXR
Notes: Standard errors in ( ) and t-statistics in [ ] Source: Eviews 6.0 outcomes
Based on the result drawn from the table 4.7, the cointegrating vector, which represents the long run equilibrium relationship among the tested variables, can be written as follows: β' = [ 1.00000 – 0.84405 – 0.41663 0.209796 – 0.155164]
The cointegrating vector coefficients for VNI (normalized to one), IP, IR, CPI, and EXR illustrate the long-run relationship between stock prices and macroeconomic variables This relationship can be represented by a specific equation, with standard errors and t-statistics indicated in round brackets and square brackets, respectively.
LVNI = 0.844050 *LIP + 0.416630 *LIR – 0.209796 *LCPI + 0.155164 *LEXR (4.2)
This study employed t-statistics to calculate the p-values of independent variable coefficients using the TDIST function in Excel The TDIST function is defined as TDIST(x, k, n), where x represents the estimated t-statistic, k denotes the degrees of freedom (calculated as the number of observations minus the number of independent variables), and n indicates the number of tails The t-statistics for the variables VNI, IP, IR, CPI, and EXR were -2.95446, -2.71998, 4.18849, and -1.10489, respectively, with a total of 61 degrees of freedom and two tails The computed p-values for the independent variable coefficients are summarized in Table 4.8.
Table 4.8 Result of long run relationship
Variables Coefficient Std Error t-Statistic Prob
Note: Dependent variable Sources: Eviews 6.0 outcomes and Excel
Table 4.8 indicates that the coefficients for industrial production and interest rate are statistically significant at the 1% level (p < 0.01), while the coefficient for the Consumer Price Index (CPI) is significant at the 0.1% level (p < 0.001) In contrast, the coefficient for the exchange rate (EXR) does not show statistical significance (p > 0.05).
The industrial production index significantly influences stock prices, with a 1% increase in the index leading to an approximate 0.8% rise in stock prices This relationship is attributed to the expectation of higher future cash flows and corporate profitability, prompting investors to purchase stocks at elevated prices This finding aligns with previous research, including studies by Hussainey and Ngoc (2009) in Vietnam and another study in Turkey (2005).
In regard to interest rate, it also had a positive impact on stock price There was an increase in stock price by 0.42%, when interest rates went up by 1%
Interest rates have a positive impact on stock prices in the Vietnamese market, as investors do not view them as viable alternative investment options Instead, they are adequately rewarded with stock returns that compensate for interest rate fluctuations.
An increase in interest rates did not lead to a decrease in stock investments, resulting in a rise in stock prices While the positive coefficient of interest rates contradicts traditional theories, it aligns with previous studies, including those by Hussainey and Ngoc (2009) for Vietnam and Wongbangpo and Sharma (2002) for Indonesia and Malaysia.
An increase in the Consumer Price Index (CPI) negatively impacts stock prices over the long term, with a 1% rise in CPI leading to a decline of approximately 0.21% in stock prices This decline is attributed to higher inflation, which raises production costs and subsequently reduces corporate profitability As companies report lower profits, their stocks become less appealing to investors seeking better returns, prompting them to sell their shares in favor of more profitable assets This trend aligns with existing theories and previous research, such as the findings of El-Nader and Alraimony (2012) in Jordan.
Wongbangpo and Sharma (2002) for Indonesia, Malaysia, Philippines, Singapore and Thailand
The exchange rate showed a positive association with stock prices; however, this finding was not statistically significant (p > 0.05), indicating insufficient evidence to assert that the exchange rate impacts stock prices In Vietnam, the State Bank controls the exchange rate, maintaining a fixed policy that results in stability Consequently, investors do not view exchange rate fluctuations as a significant risk factor, which explains the minimal influence of the exchange rate on stock prices.
As the equation (4.1) shows, in order to examine the short run relationship between stock price and macroeconomic variables, the error correction term
(ECT t-1 ), which is the lag of the residuals generated from the cointegaring equation
(4.2), was included in the VECM equation Since the computed optimal lag length was one (m = 1), the lag length used in VECM specification was also one
The VECM equation (4.1) has been reformulated for clarity and precision For the latest updates and resources, please refer to the provided email address.
This study utilized the Ordinary Least Square (OLS) method in order to estimate the short run coefficients in equation (4.3) The result of VECM was presented in the table 4.9
Table 4.9 Result of short run relationship
Variables Coefficient Std Error t-Statistic Prob
Adjusted R-squared 0.301020 S.D dependent var 0.101473 S.E of regression 0.084837 Akaike info criterion -1.991740 Sum squared resid 0.403046 Schwarz criterion -1.753614 Log likelihood 69.73982 Hannan-Quinn criter -1.898084
Note: Dependent variable Source: Eviews 6.0 outcomes
Based on the findings from Table 4.9, the short-run relationship between stock prices and macroeconomic variables, as represented by Equation (4.3), has been reformulated.
In the short run, the stock price is positively influenced by its own previous value and the interest rate with a one-month lag, indicating that investors take time to adjust their portfolios in response to changes Specifically, the previous stock price index significantly impacts the current index, with a coefficient of 0.25 This means that a 1% increase in the previous stock price index leads to a 0.25% rise in the current stock price index, as investors anticipate further increases and subsequently buy more stocks, driving the price higher.
There is a notable positive correlation between interest rates and stock prices in the short term, with a 1% increase in interest rates leading to a 0.33% rise in stock prices This relationship exists because investors in the Vietnamese stock market do not view interest rates as viable alternative investment options, and they receive adequate compensation through stock returns.
An increase in interest rates did not lead to a decrease in stock investments, resulting in a rise in stock prices.
Diagnostic tests
This study performed a series of diagnostic tests, including the serial correlation LM test, normality test, and heteroskedasticity test, to assess the consistency of the model with the data.
H 0 : There is no autocorrelation problem
H 1 : There is an autocorrelation problem
Breusch-Godfrey Serial Correlation LM Test:
Obs*R-squared 0.874237 Prob Chi-Square(2) 0.6459
The Durbin-Watson test indicated no first-order autocorrelation, with a value of 1.97, which is close to two Additionally, the serial correlation LM test showed no evidence of higher-order autocorrelation in the residuals, as both the F-statistic and Chi-squared p-values were greater than 0.05, supporting the null hypothesis of no autocorrelation Furthermore, the correlogram Q-statistic test confirmed the absence of autocorrelation issues (refer to table A17).
Therefore, the results of these tests indicate that there was no autocorrelation problem in the residual
H 1 : Residuals are not normally distributed
Table 4.12 Result of normality test
The normality test revealed that the residuals followed a normal distribution, as indicated by a Jarque-Bera p-value of 0.95, which exceeds the 0.05 threshold This result suggests that we cannot reject the null hypothesis of normal distribution for the residuals.
Series: VECM Residuals Sample 2008M03 2013M05 Observations 63
Mean -0.003421 Median -0.001043 Maximum 0.179373 Minimum -0.203587 Std Dev 0.080635 Skewness -0.050110 Kurtosis 2.821986
Jarque-Bera 0.109549 Probability 0.946699 tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
H 0 : There is no heteroskedasticity problem
Heteroskedasticity Test: Breusch-Pagan-Godfrey
Obs*R-squared 12.73388 Prob Chi-Square(10) 0.2389 Scaled explained SS 9.543940 Prob Chi-Square(10) 0.4814
The heteroskedasticity test indicated that there was no issue with heteroskedasticity in the residuals, as the p-value of the F-statistic was 0.24, which is greater than the 0.05 threshold Therefore, the null hypothesis of no heteroskedasticity could not be rejected.