Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1598-7248 (Print)
ISSN : 2234-6473 (Online)
Industrial Engineering & Management Systems Vol.21 No.1 pp.85-109
DOI : https://doi.org/10.7232/iems.2022.21.1.085

Stock Returns Response to Internal and External Shocks during the COVID-19 Pandemic in Indonesia: A Comparison Study

Rossanto Dwi Handoyo*, Kabiru Hannafi Ibrahim, Frandy Yosza Indrawan
Faculty of Economics and Business, Universitas Airlangga, Surabaya, Indonesia
Faculty of Social and Management Sciences, Federal University, Birnin Kebbi, Nigeria
Faculty of Economics and Business, Universitas Airlangga, Surabaya, Indonesia
*Corresponding Author, E-mail: rossanto_dh@feb.unair.ac.id
June 8, 2021 October 3, 2021 December 7, 2021

ABSTRACT


This study uses impulse response function and variance decomposition from the Vector Error Correction Model (VECM) and analyses the response of Jakarta Composite Index (JCI) return to the shocks of oil price, gold price, the exchange rate, interbank interest rates, COVID-19 cases, and the stock market index of Malaysia, Singapore, Thailand, Japan, and the United States. Daily secondary data were used for the analysis and our empirical strategy from the impulse response function divulges that JCI return responds positively to the shock of the Malaysia and Japan stock indices and negatively to Singapore, Thailand, and the United States stock indices. Our finding further reveals that JCI return responds positively to its shock, shocks of the gold price, exchange rates, and negatively respond to the shocks of oil price, interbank rate, and COVID-19 cases. Therefore, based on the study findings policy recommendations are made to mitigate the negative influence of the shocks variables on JCI return.



초록


    1. INTRODUCTION

    The Jakarta composite index (JCI) is simply the Jakarta price index which was initiated on the 10th August 1982 with a 100 index value as its base. The index is an improved capitalisation-weighted index of stocks listed in the Indonesian stock exchange market. The JCI is one of the indices listed in the Indonesian stock exchange and it trails the recital of companies in the Indonesian stock exchange. The index consists of all issuers with initial public offerings (IPOs) in the Indonesian stock exchange. The stock market is often construed to be one of the necessary financial institutions in the economy. It is one of the considerations to predict and determine economic conditions (Bissoon et al., 2016). In addition, it has a vital role in economic growth (Khan et al., 2017a). A view from Dong and Yoon (2019) is that the stock market can reflect actual economic activity and it is much more predictable despite its tentative nature of fluctuations as pointed out by Faghihi-Nezhad and Minaei-Bidgoli (2018) using AI-based models. For instance, a more recent work by Celebi and Hönig (2019), Demir (2019), Gopinathan and Durai (2019), Hussain et al. (2012), (Khan and Khan (2018), Lawal et al. (2018), Aigbovo and Izekor (2015), and Wahyudi et al. (2017) among others has found a strong nexus between macroeconomic variables and the stock exchange market.

    The coronavirus illness was named COVID-19 by the World Health Organization (WHO) (Apornak, 2021). Its outbreak started in December 2019 and has resulted in a global crisis with a negative effect both socially and economically (Giones et al., 2020). The COVID-19 pandemic has weakened the Indonesian economy. Since its inception, the Rupiah exchange rate has weakened against the US dollar. The rupiah exchange rate has fallen to the lowest rate on March 24, 2020, exchanged for IDR 16,504.80 per US dollar. In addition, the COVID-19 pandemic has crippled the stock market. The Jakarta Composite Index (JCI) has experienced a down-trend since the beginning of March 2020. The movement of JCI has experienced its lowest point on March 24, 2020, with a value of 3,937.63 which later slowly increased until the end of August 2020, which stood at 5,149.63. The Indonesian stock exchange market experienced a temporary trade freeze on March 12, 2020. The trade freeze was caused by a decrease in the JCI which reached 5.01 percent. Freeze trading also occurred precisely on March 13, 17, 19, 23, and 30, 2020. The JCI began to fall on March 9, 2020, by 6 percent, but there were no rules, so trading transactions were not stopped (Sadono, 2020). This trading freeze indicates that investors often sell shares on the Indonesian stock exchange.

    Foreign and domestic investors investing in the Indonesia Stock Exchange experienced a decline from January 2020 to March 2020. In April 2020, foreign and domestic investors experienced an increase compared to the previous months. Foreign and domestic investors, which fell quite sharply in March 2020, indicate an increase in selling from within and outside the country. The composition of foreign investors in Indonesia prevailed more than domestic investors from January to March 2020 while foreign investors prevailed more than domestic investors from April to August 2020. The composition of these investors shows that stocks in Indonesia are vulnerable to external and internal selling.

    In Indonesia, external and internal factors are used as considerations for investing in shares. Therefore, this study was carried out by raising the economic problems that have occurred due to the rapid growth in cases of Coronavirus Disease (COVID-19). Previous studies mainly focused on the effects of the COVID-19 on stock market conditions (Liu et al., 2020;Mishra, 2020). Studies linking the shock of COVID-19 and other factors to the stock market are still rare. Previous studies on factors influencing returns in the stock market were essentially based on macroeconomic variables and other commodity prices (Contuk et al., 2013;Bissoon et al., 2016;Khan et al., 2019;Khan, 2019). In addition, studies on the impact of COVID-19 on the JCI return are still rare. Therefore, this study aims to bridge this gap in the literature by examining the stock returns’ response to internal and external shocks during the COVID-19 Pandemic. The study used impulse response function and variance decomposition from the Vector Error Correction Model (VECM) to determine the response of JCI return to the shocks of world crude oil prices, world gold prices, exchange rate, interbank interest rates, the addition of the COVID-19 cases and stock prices of other countries.

    The remaining parts of this paper are structured as follows; in section two we present the review of external and internal economic factors, in section three we present the literature review, in section four we present the data and methodology, in section five we present the study findings and the analysis, while in the final section six we conclude the paper and offer policy suggestions.

    2. REVIEW OF EXTERNAL AND INTERNAL ECONOMIC FACTORS

    In this section of the study we review some internal and external factors as follows;

    2.1 Stock Market Index

    According to Jazairi et al. (2006: 8312), the stock price index is the value of the movement of shares or other financial assets displayed or indicated by the time being updated. The indices are evaluated by the time of day, several times a day, and are more prone to vulnerability. Rauterberg and Verstein (2013) explain the benefits of the stock index for investors to analyse the impact of the economic situation. Another benefit of the stock index is that it reduces investment costs by not using an investment manager. Financial ratios have been found to be important predictors of changes in the stock prices which also reduce the effect of macroeconomic variables on the stock market returns (Kwag and Kim, 2013).

    2.2 Oil Price

    Fattouh (2011) explains that petroleum is not the same or single commodity. Crude oil is traded internationally and has a variety of different types, qualities, and characteristics. There are benchmarks for setting petroleum prices globally which is the centre of the overall oil price information system. According to Carollo (2012), the oil price does not solely reflect the balance of market demand and supply prices. The relationship between oil prices and the stock market is reflected by the company’s profit. The rising oil price will increase the cost of productive inputs. Investors will sell shares if the company’s profit decreases (Masih et al., 2011). The increase in oil prices also impacts inflation and causes a decrease in investor expectations from the stock market (Youssef and Mokni, 2019). The oil price shock has a positive effect on the stock market return of oil-exporting countries. Meanwhile, it negatively responds to the stock market return of oil-importing countries (Park and Ratti, 2008).

    2.3 Gold Price

    Aggarwal and Lucey (2005) explain that gold is a type of metal traded 24 hours a day that is vital for human activities and also served as an investment mechanism. Gold can be a safe store and hedge of value in times of global economic uncertainty (Baur and McDermott, 2010). The link between the gold price and economic conditions is observed during the economic crisis. The economic crisis prompted investors to seek safe-haven assets such as gold (Markwat et al., 2009). Gold investment is considered as a means of protection against inflation when the stock market is unstable (Baur and Lucey, 2010). The price of gold can be used to indicate the economy and market disruptions (Contuk et al., 2013). Gold as an international investment mechanism causes stable demand and tends to increase every day (Narang and Singh, 2012).

    2.4 Currency Exchange Rates

    Salvatore (2013: 429) explains that the exchange rate is the country’s currency needed to buy another country’s currency. Exchange rates can appreciate or depreciate depending on the market situation. Mankiw (2016: 155) divides the exchange rate into the nominal and the real exchange rate. The nominal rate is the relative price of the two countries’ currencies while the actual rate is the relative price of goods traded by two countries. The actual exchange rate is expressed in Equation (1) as follows;

    R e a l   E x c h a n g e   R a t e = E x c h a n g e   R a t e * P r i c e   o f   D o m e s t i c   G o o d s P r i c e   o f   F o r e i g n   G o o d s
    (1)

    Currency exchange rates can affect market conditions. It can affect company earnings Šimáková (2017) because the exchange rate movements will impact the company’s shares. Besides, depreciation of currency can have a positive impact by increasing exports due to lower prices. An increase in sales of goods will increase company profits, which creates a positive sentiment for stock investors (Paramati and Gupta, 2013).

    2.5 Interbank Interest Rates

    According to Mishkin (2012: 38), the interest rate is the cost of borrowing or the price paid for leasing funds and money. Interest rates vary in the economy, depending on liquidity and the safety of the accompanying risks. The Indonesia Overnight Index Average (IndONIA) is an interest rates index from lending and lending transactions in rupiah currency that does not use overnight interbank collateral. The IndONIA is determined based on reports of all banks to the Bank of Indonesia. The interest rate is important in determining the share price. Investors make changes in interest rates as an indicator of share price expectations (Amarasinghe, 2015). Changes in interest rate policies can affect share prices. The market defines the comprehensive policy as an economic boost that positively impacts stock investors’ sentiment (Bissoon et al., 2016).

    2.6 Disease Incidence

    A pandemic is an epidemic that has spread in several countries or continents and affects a large proportion of the human population. The COVID-19 pandemic provides information to investors, policy administrators, and society as a whole that will impact economic disruption (Goodell, 2020). Stock investors consider the COVID-19 virus as an obstacle to economic activity and sell shares (Liu et al., 2020). News of the growth of the COVID-19 cases also hurts the return of the stock market concerned (Al-awadhi et al., 2020). In addition to having a negative effect, COVID-19 causes an increase in the shares of companies in the pharmaceutical and biotechnology sectors (Ozdurak et al., 2020).

    2.7 The Influence of Stock Markets from Other Countries

    The modern markets have made it easier for global investors to access the stock markets of other countries. Ease of access allows for easy flow of information and the integration of countries’ stock markets to the global stock market (Arouri and Foulquier, 2012). Stock market integration between countries creates the risk of shifting shocks from rising and falling share prices (Shen, 2018). Stock market integration between countries can occur in regional areas such as Southeast Asia (Thao and Daly, 2012). Stock market integration can also occur with members of organizations such as the Gulf Cooperation Council (GCC) or state cooperation in the Arabian Gulf region (Sbeiti and Alshammari, 2010). In addition, the stock market in the United States can affect the stock market in the Asian region (Lee and Chou, 2019). The size and influence of other countries’ stock markets can be due to global economic disruptions (Wu, 2020).

    3. LITERATURE REVIEW

    An empirical study by Gokmenoglu and Fazlollahi (2015) observed that the price of Brent oil and London gold hurts the S&P 500 stock index in the United States stock market. Rahman and Mustafa (2018) also revealed that in the short term, oil and gold prices negatively affect the returns of Standard & Poor’s 500 stock index. While in the long run, oil prices assert a negative effect, and gold prices have a positive effect. A similar study by Khan et al. (2019) shows that in both the short and long term the increase in oil prices in West Texas Intermediate (WTI) harms the return of the Shanghai stock index. While examining the response of stocks and oil prices to the shocks before and during COVID-19, Salisu et al. (2020) applied panel vector autoregressive and observed that during the pandemic stock and oil markets experience higher impact than before the global pandemic. This finding has been further supported by the result from the logit model.

    A study by Robiyanto (2018) indicates that exchange rate and interest rates harm the return of JCI, while the price of gold shows a positive effect. Another study by Nordin et al. (2014) indicates that palm oil prices, interest rates, and currency exchange rates significantly and negatively affect the Malaysian stock market index. Giri and Joshi (2017) also revealed that interest rates and exchange rates affect the Sensex index positively in the long and short term, while the crude oil price negatively affects the Sensex index.

    Darmawan et al.’s (2020) findings indicate that the JCI responds negatively to changes in the shock of Brent oil prices and the Japanese stock index. Furthermore, JCI responds to the shocks of Singapore, China, the United States, Britain, and Germany. In another study, Octavia and Wijaya (2020) show that the stock index of Malaysia and Japan has a positive nexus with the JCI, while the stock index of the United States has a negative nexus with the JCI. Similarly, Thao and Daly (2012) showed a one-way and two-way nexus between the Philippines, Malaysia, Singapore, Indonesia, Thailand stock markets which was not found in the case of the Vietnam stock market.

    Lee and Chou’s (2019) finding indicates that the nexus between the United States, China, Hong Kong, Taiwan, Japan, Indonesia, Philippines, Malaysia, Thailand, and Singapore’s daily stock market returns occurs due to global events and crises. A similar study by Wu (2020) revealed that the Japanese stock index strongly influences Southeast Asian stock markets during global economic disruptions. In the case of D8 countries, Ranjbar et al. (2018) have found macroeconomic variables to assert a positive impact on stock returns.

    A study by Al-Awadhi et al. (2020) reports that the growth in the daily cases of COVID-19 and confirmed cases of death harmed the returns of company stock on the Hang Seng index and Shanghai. A finding by Liu et al. (2020) observed the impact of the COVID-19 outbreak on the return of 21 stock indexes in countries affected by the outbreak, including Indonesia. Another study by Ozdurak et al. (2020) shows that the stocks of pharmaceutical and biotechnology companies traded on the NYSE exchange show a positive value along with the outbreak of COVID- 19. Similarly, Topcu and Gulal’s (2020) study finding shows that an increase in the exchange rate, shock in oil price, and COVID-19 cases harm the stock market. In the case of South Korea and the United States, Wang and Park’s (2021) empirical strategy revealed that South Korea’s financial market is more susceptible to the COVID- 19 pandemic while for the United States it is found that only the stock market is negatively affected by the growing number of COVID-19 cases. Revised exchange rate as a factor influencing stock returns has been found to influence the fair value of assets during COVID-19 in Iraq’s economy (Ali et al., 2020).

    4. DATA AND METHODOLOGY OF THE STUDY

    4.1 Description of the Data

    Daily secondary data are collected for this study with five days data series per week starting from COVID-19 cases in Indonesia. The empirical analysis was based on 354 observations over the period March 2, 2020, to September 1, 2021. The analysed variables include; the world oil price (Oil), world gold price (Gold), the rupiah exchange rate (XR), the cases of COVID-19 per day (COVID), the stock price index of the Southeast Asia region consisting of the Jakarta Composite Index (JCI), Malaysia (KLSE), Singapore (STI), Thailand (SETI), and the East Asia region represented by Japan (Nikkei) and the United States (DJIA) and the exception of the interbank interest rate variable (IB). Impulse response analysis and variance decomposition from the VECM were applied for the analysis of the response of JCI return to the shocks from world oil prices, world gold prices, the rupiah exchange rate against the US dollar, interbank interest rates, the number of COVID-19 cases in Indonesia and stock price indexes of other countries. The data were obtained from various sources like Bank of Indonesia, Bloomberg, U.S Energy Information Administration (EIA), and yahoo finance.

    4.2 Analytical Models

    Following a study by Tripathy (2015) in this study we first looked at the daily JCI return which is expressed in Equation (2) as follows:

    R t = l n ( P t P t 1 )   × 100
    (2)

    where Rt is the return in period t, Pt is the current price, and Pt-1 is the previous price in natural logarithms. Gujarati and Porter (2008: 162) show the stages of the natural logarithm, which are shown as follows:

    Y t =   Y 0 ( 1 + r ) t
    (3)

    By going through the natural logarithm steps, it becomes

    l n Y t =   l n Y 0 + t   l n ( 1 + r )
    (4)

    where:

    • Yt : Denote real expenditure on services at time t

    • Y0 : Denote the initial value of expenditure on services

    • r : Denote the compound rate of growth of the real expenditure Y.

    The VECM used in this study was adopted following a study by Singh and Sharma (2018) which is divided into two models. Model one consists of variables which include; oil price (Oil), gold price (Gold), rupiah exchange rate (XR), interbank interest rate (IB), and additional cases of COVID-19 (COVID). Model two consists of the stock market index variables of Malaysia (KLSE), Singapore (STI), Thailand (SETI), Japan (Nikkei), and the United States (DJIA). Model 1 is expressed in the following specification:

    Δ L N R T J C I t =   α +   β E C T t 1 +   i = 1 k α i Δ L N R T J C I t i +   t = 1 k b i Δ L N O i l t i + t = 1 k γ i Δ L N G o l d t i + t = 1 k Ф i Δ X R t i + t = 1 k b i Δ L N I B t i   + t = 1 k Ω i Δ L N C O V I D t i + ε t
    (5)

    where:

    • α : is the constant term

    • ECT: is the error correction term

    • β, αi, bi, γi, Фi, Ψi, Ωi : are the parameters of the explanatory variables to be estimated

    • i and k: are the lag length

    • ε : is the classical error term

    • RTJCI: is the return of the Jakarta composite stockprice index

    • Oil: is the world oil price

    • Gold: is the world gold price

    • XR: is the exchange rate

    • IB: is the interbank rate

    • COVID: is the addition of daily COVID-19 cases.

    • Model 2 is expressed in the following specification:

    Δ L N R T J C I t =   α +   β E C T t 1 + i = 1 k α i Δ L N R T J C I t i +   t = 1 k b i Δ L N K L S E t i + t = 1 k b i Δ L N S T I t i + t = 1 k b i Δ L N S E T I t i + t = 1 k b i Δ L N N i k k e i t i   + t = 1 k b i Δ L N D J I A t i   +   ε t
    (6)

    where:

    • α, ECT, and RTJCI: are as defined in Equation (5)

    • KLSE : is the Malaysian stock index

    • STI : is the Singapore stock index

    • SETI : is the Thailand stock index

    • Nikkei : is the Japan stock index

    • DJIA : is the United States stock index.

    Except for the interbank interest rate, all the variables in Equations (5) and (6) are transformed into a natural logarithm. This is important to normalize the data and to obtain the model parameters as elasticities for easy interpretation and to reduce the influence of extreme values from our data set.

    4.3 The Definition of Operational Variables

    This section shows the meaning of each variable used in this study and provides its defined limit as follows:

    4.3.1 Jakarta Composite Index

    It is an index that shows the price based on the spot market closing of the Indonesian stock market index with a return stage. Other countries’ stock index variables consist of Malaysia, Singapore, Thailand, Japan, and the United States. The study data covered the period March 2020 to September 2021.

    4.3.2 Oil Price

    The price of crude oil is based on West Texas Intermediate (WTI) oil in US dollars per barrel from March 2020 to September 2021.

    4.3.3 Gold Price

    The gold price is based on the closing market price. Spot in US dollars per Troy ounce from the period March 2020 to September 2021.

    4.3.4 Exchange Rate of Rupiah against US Dollars

    The exchange rate is based on the currency trading mechanism indicated by the closing value of spot market trading from March 2020 to September 2021.

    4.3.5 The Interbank Interest Rates

    IndONIA interest rate is a report of all banks to the bank of Indonesia, which is sourced from the average interest rate for borrowing and lending transactions in rupiah that does not use overnight collateral in percentage from March 2020 to September 2021.

    4.3.6 Increase in COVID-19 Cases

    Data of the new COVID-19 cases announced by the government from March 2, 2020, to September 2021.

    4.4 Analysis Technique

    Vector Error Correction Model (VECM)

    A VECM is based on an estimate of Vector Auto Regression (VAR). It allows for non-stationary data at a level but has cointegration and thus indicating a long-term relationship by allowing the dynamics of a short-term relationship (Tripathi and Kumar, 2016). VECM is used for non-stationary data while differencing “d” all the variables in the model (Ali and Sun, 2017). The steps to determine the VECM method are as follows:

    • Step 1: Stationarity Test

    The stationarity test is used to determine whether the statistical properties of the time series do change over time. In this study, the Augmented Dickey-Fuller (ADF) test for stationarity was applied. Data that is not stationary at the level will be followed by the first difference, with the following hypothesis:

    • H0 : P-value > 0.01, 0.05, 0.1 indicates nonstationary data.

    • H1 : P-value < 0.01, 0.05, 0.1 indicates stationary data.

    Step 2: Lag Test

    The Lag Test is used to obtain the VECM estimation. In determining the optimal lag length, the study used Akaike Information Criterion (AIC), Schwarz Information Criterion (SIC), and Hannan-Quinn Information (HQ) Criterion.

    • Step 3: Cointegration Test

    A cointegration test for variables is used to see if there is a long-term relationship between variables in the study (Narang and Singh, 2012). This can be determined by comparing the trace statistical value and the critical value at the 5% significance level. The long-term relationship between variables occurs when the trace statistic value is greater than the critical value at the 5% significance level. The trace statistics are shown as follows:

    τ t r a c e =   T   l o g   ( 1 λ i )
    (7)

    where λ is the Eigenvalue and T is the number of observations.

    Vector Error Correction Model (VECM) Analysis

    VECM analysis is used to determine models that show long and short-term effects. The influence is indicated by the t statistical value greater than the t table value. The coefficient value of each variable indicates the magnitude of the influence value of the variable. This stage is a process for impulse response analysis and variance decomposition.

    Impulse Response Function and Variance Decomposition Analysis

    The impulse response function is used to determine the variable response to its shock and the shock from other variables in the system (Tripathy, 2015). Enders (2014: 285) shows how the shock affects the variables as described below:

    • Step 4: VAR Analysis

      y t = b 10 b 12 z t + γ 11 y t 1 + γ 12 z t 1 + ε y t
      (8)

      z t =   b 20 b 21 y t + γ 21 y t 1 + γ 22 z t 1 + ε z t
      (9)

    Equations (8) and (9) are examples of two variables that have a simultaneous relationship. Equations (8) and (9) are non-reduced forms of the equation, where the yt variable influences the zt variable and the variable zt influences on the yt . The matrix used to transform equations (8) and (9) is expressed as follows:

    [ 1 b 12 b 21 1 ] [ y t z t ] =   [ b 10 b 20 ] +   [ γ 11   γ 12 γ 21 γ 22 ]   [ y t 1 z t 1 ] +   [ ε y t ε z t ]
    (10)

    or

    B x t = Γ 0 + Γ 1 x t 1 + ε t
    (11)

    where:

    = [ 1 b 12 b 21 1 ] , x t =   [ y t z t ] ,   Γ 0 =   [ b 10 b 20 ] , Γ 1 =   [ γ 11   γ 12 γ 21 γ 22 ] ,   ε t =   [ ε y t ε z t ]
    (12)

    The standard VAR model can be obtained from the product of the second segment B−1 , shown as follows:

    x t = A 0 + A 1 x t 1 + e t
    (13)

    where

    A 0   = B 1 Γ 0 , A 1 = B 1 Γ 1 , e t = B 1 ε t
    (14)

    Based on this description, it can be explained that ai0 is the i element of the vector A0, aij is the-i row element of the jth column of the matrix A1 and eit is the ith element of the et vector. Based on this explanation, equation (13) is rewritten as follows:

    y t = a 10 a 11 y t 1 + a 12 z t 1 + e 1 t
    (15)

    z t = a 20 a 21 y t 1 + a 22 z t 1 + e 2 t
    (16)

    Equations (8) and (9) against (15) and (16) have different views. Equations (8) and (9) are the structures of the primitive VAR. Equations (15) and (16) are VAR structures with standard forms. The error is shown by e 1 t and e 2 t is the combined shock of εyt and εzt . Based on et = B−1εt , error e 1 t and e 2 t can be shown as follows:

    e 1 t = ( ε y t b 12 ε z t ) ( 1 b 12 b 21 )
    (17)

    e 2 t = ( ε z t b 21 ε y t ) ( 1 b 12 b 21 )
    (18)

    Error e 1 t and e 2 t are known to have zero mean, constant variance, and not individually correlated. This error exists because the εyt and εzt shocks are whitenoise mechanisms. The character of e 1 t can be found using equation (17) which is shown as follows:

    E e 1 t = E ( ε y t b 12 ε z t ) ( 1 b 12 b 21 ) = 0
    (19)

    The variance of e 1 t derives from

    E e 1 t 2 = E ( ε y t b 12 ε z t ) ( 1 b 12 b 21 ) 2 = ( σ y 2 +   b 12 2 σ z 2 ) ( 1 b 12 b 21 ) 2
    (20)

    Autocorrelation of e 1 t e 1 t i   is shown as follows:

    E e 1 t e 1 t i   = E [ ( ε y t b 12 ε z t ) ( ε y t i b 12 ε z t i ) ] ( 1 b 12 b 21 ) 2 = 0
    (21)

    Correlation of e 1 t and e 2 t is shown as follows:

    E e 1 t e 2 t =   E [ ( ε y t b 12 ε z t ) ( ε y t i b 21 ε z t ) ] ( 1 b 12 b 21 ) 2 =   ( b 21 σ y 2 +   b 12 σ z 2 ) ( 1 b 12 b 21 ) 2
    (22)

    Equation (22) is rarely zero or shows a correlation between the two shocks. Equation (22) can show that the shock is uncorrelated or has a zero value because yt to zt and zt to yt don’t have any effect. The steps to explain the shock matrix e 1 t and e 2 t are shown as follows:

    = [ v a r ( e 1 t ) c o v ( e 1 t , e 2 t ) c o v ( e 1 t , e 2 t ) v a r ( e 2 t ) ]
    (23)

    The simple form of equation (22) is shown as follows:

    = [ σ 1 2 σ 12 σ 21 σ 2 2 ]
    (24)

    where v a r ( e 1 t ) = σ i 2 and c o v ( e 1 t , e 2 t ) = σ 12 = σ 21

    • Step 5:> Stability and Stationarity

    The requirement for stability from an autoregressive model y t =   a 0 +   a 1 y t 1 +   ε t is a1 smaller than the unit of absolute value. In addition, the condition is that there is a direct analogy to the equation (13). The brute force method is used to solve equation (13) iterated as follows:

    x t =   A 0 +   A 1 ( A 0 +   A 1 x t 2 +   e t 1 ) + e t = ( I + A 1 ) A 0 +   A 1 2 x t 2 + A 1 e t 1 + e t
    (25)

    where I = 2 x 2 identity matrix.

    After n iterations are shown as follows:

    x t = ( I + A 1 + ... + A 1 n )   A 0 + i = 0 n A 1 i e t i +   A 1 n + 1 x t n 1
    (26)

    The effect of the continued iteration is that A 1 n will disappear into infinite approximations. If stability conditions have been achieved, then xt can be written as follows:

    x t =   μ +   i = 0 A 1 i e t i
    (27)

    where μ = [ y ¯ z ¯ ]   '

    y ¯ =   [ a 10 ( 1 a 22 ) +   a 12 a 20 ] Δ
    (28)

    Δ   = ( 1 a 11 ) ( 1 a 22 ) a 12 a 21
    (29)

    z ¯ =   [ a 20 ( 1 a 11 ) +   a 21 a 10 ] Δ
    (30)

    • Step 6: Estimation and Identification

    A theory that assumes variable Z to affect variable Y, while variable Y cannot affect variable Z. To analyse the theory, the value of b21 will be estimated with 0 or b21 = 0 and shown by equations (31) and (32) as follows:

    y t = b 10 b 12 z t + γ 11 y t 1 + γ 12 z t 1 + ε y t
    (31)

    z t = b 20 + γ 21 y t 1 + γ 22 z t 1 + ε z t
    (32)

    The restriction will show B−1 as follows:

    B 1 =   [ 1 b 12 0 1 ]

    The result of primitive VAR and B−1 is shown as follows:

    [ y t z t ] =   [ 1 b 12 0 1 ]   [ b 10 b 20 ] +   [ 1 b 12 0 1 ]   [ γ 11   γ 12 γ 21 γ 22 ]   [ y t 1 z t 1 ] +   [ 1 b 12 0 1 ]   [ ε y t ε z t ] [ y t z t ] = [ b 10 b 12 b 20 b 20 ] +   [ γ 11 b 12 γ 21 γ 12   b 12 γ 22 γ 21 γ 22 ]   [ y t 1 z t 1 ] +   [ ε y t b 12 ε z t ε z t ]
    (33)

    Equations (15) and (16) are used to estimate and produce the following parameters:

    a 10 =   b 10 b 12 b 20 a 11 =   γ 11 b 12 γ 21 a 12 =   γ 12   b 12 γ 22 a 20 =   b 20       a 21 =   γ 21 a 21 = γ 22

    The restriction b21=0 results in e 1 t = ε y t b 12 ε z t and e 2 t = ε z t , then

    v a r ( e 1 ) =   σ y 2 +   b 12 2 σ z 2 v a r ( e 2 ) =   σ z 2 c o v ( e 1 , e 2 ) = b 12 σ z 2

    The estimates of VAR parameters are shown as;

    a 10 , a 11 , a 12 , a 20 , a 21 , a 22 , v a r ( e 1 ) , v a r ( e 2 ) ,   c o v ( e 1 , e 2 )

    Are substituted into equations (31) and (32) to solve for b 10 , b 12 , γ 11 , γ 12 , b 20 , γ 21 , y 22 , σ y 2 and σ z 2 .

    • Step 7: Impulse Response

    Equations (31) and (32) can be converted into a matrix form as follows:

    [ y t z t ] = [ a 10 a 20 ] + [ a 11 a 12 a 21 a 22 ]   [ y t 1 z t 1 ] + [ e 1 t e 2 t ]
    (34)

    Equation (34) can be shown by equation (27) as follows:

    [ y t z t ] =   [ y ¯ z ¯ ] +   i = 0 [ a 11 a 12 a 21 a 22 ] i   [ e 1 t i e 2 t i ]
    (35)

    The vectors of the error equations (17) and (18) are shown as follows:

    [ e 1 t e 2 t ] =   1 1   b 12 b 21   [ 1 b 12 b 21 1 ]   [ ε y t ε z t ]
    (36)

    Equations (35) and (36) are combined as follows:

    [ y t z t ] =   [ y ¯ z ¯ ] + 1 1   b 12 b 21   i = 0 [ a 11 a 12 a 21 a 22 ] i   [ 1 b 12 b 21 1 ]     [ ε y t i ε z t i ]
    (37)

    The ϕi matrix is used to simplify the explanation shown as follows:

    ϕ i =   A 1 i 1   b 12 b 21   [ 1 b 12 b 21 1 ]
    (38)

    Equation (37) with the matrix ϕi can be shown as follows:

    [ y t z t ] =   [ y ¯ z ¯ ] + i = 0 [ ϕ 11 ( i ) ϕ 12 ( i ) ϕ 21 ( i ) ϕ 22 ( i ) ]   [ ε y t i ε z t i ]
    (39)

    Equation (39) can be clarified by

    x t = μ + i = 0 ϕ i ε t i
    (40)

    The impulse response function is indicated by the ϕ 11 ( i ) , ϕ 12 ( i ) , ϕ 21 ( i ) , ϕ 22 ( i ) coefficient. The shock effect of εyt and εzt on yt and zt can be seen from the coefficients ϕ 11 ( i ) , ϕ 12 ( i ) , ϕ 21 ( i ) , ϕ 22 ( i ) . The coefficient ϕ12 (0) is the response to the change in εyt to yt . The same thing is shown by ϕ11(1) and ϕ12(1) which is the response of the changes in εyt−1 and εzt−1 to yt . The VAR system in general has been described as underidentified, where the impulse response estimation cannot be done. Restrictions are used so that variables can be estimated with an impulse response or in just identified conditions. The restriction b21=0 will result in a residual of e 1 t =   ε y t b 12 ε z t . Cholesky decomposition is used for the restriction of the VAR equation. Cholesky decomposition aims at the shock of εyt not directly affecting the value of zt , on the contrary, the shock of εzt can directly affect yt and zt .

    Variance decomposition is used to show the contribution of a variable shock to other variables. Variance decomposition can show the value of the variable shock contribution (Giri and Joshi, 2017). Enders (2014: 302) explains that forecast error variance decomposition shows how much shock contributes to itself and other variables. Variance decomposition explains that if the shock of εzt cannot explain the change in yt , then yt is exogenous. The variable yt can become endogenous if the shock from εzt can explain the change in yt .

    5. RESULTS AND DISCUSSION

    5.1 VECM Model Estimation Results

    5.1.1 Stationarity Test

    The variables test of stationarity is the first step for the VECM method. An Augmented Dickey-Fuller (ADF) test for stationarity was applied to the study data. Table 1 demonstrates that only three variables (lnRTJCI, lnGold, and lnCOVID) are stationary at the level. As a result, the stationarity test at the first difference was conducted on all variables. The results of the stationarity test at first difference indicate that all variables were stationary.

    5.1.2 Test Lag

    For models 1 and 2, two tests of lag were performed. Table 2 shows that the best lag for both models 1 and 2 is lag 2 which is used to test cointegration associated with models 1 and 2.

    5.1.3 Cointegration Test

    The cointegration test results can be seen by looking at the trace statistic value and the critical value at a 5% significance level. A long-term relationship is indicated by the trace statistic value more significant than the critical value. Table 3 shows that both models 1 and 2 with lag 2 established a long-term relationship between variables.

    5.2 Analysis of Impulse Response and Variance Decomposition

    5.2.1 Analysis of Impulse Response

    Figure 1 displays the response of JCI return to; its shocks, shock of world oil prices, the shock of world gold prices, the shock of rupiah exchange rates, the shock of interbank interest rates, and the shock of COVID-19 cases. Finding from the impulse response function revealed that, except for the second period, the JCI return responds positively to its shock for the whole period under study. The JCI return responds negatively to the shock of world oil prices for the whole period under study. Except for the third period, the JCI return responds positively to the shock of world gold prices for the whole period under study. Similarly, except for the third period, the JCI return responds positively to the shock of the rupiah exchange rate for the whole period under study. Except for the fifth period, the JCI return responds negatively to the shock of the interbank interest rate for the whole period under study. Similarly, except for the fifth period, the JCI return responds negatively to the shock of the COVID-19 cases for the whole period under study.

    Figure 2 also displays the response of JCI returns to the shocks of the stock indices of Malaysia, Singapore, Thailand, Japan, and the United States. Results from the impulse response function indicate that the JCI return responds positively to the shock of the Malaysian stock index for the whole period under study. Likewise, the JCI return responds negatively to the shocks of the Singapore and the United State stock indices for the whole period under study. Except for the fourth period, the JCI return responds negatively to the shock of Thailand’s stock index and positively to the shock of the Japanese stock index for the whole period under study.

    Therefore looking at the empirical models 1 and 2 that were based on the analysis of the impulse response function as depicted in Figures 1 and 2, the Jakarta composite index responds positively and negatively to the internal and external shocks related to the Jakarta composite index, word oil prices, world gold prices, exchange rate, interbank rate, COVID-19 cases, stock indices of Malaysia, Singapore, Thailand, Japan, and the United States.

    5.2.2 Analysis of Variance Decomposition

    Table 4 shows a summary of variance decomposition results for stock return on all shock variables consisting of models 1 and 2. In model 1, the shock of JCI return determines the return of the JCI. The order of exchange rates indicates the shock contribution of other variables in explaining the return of JCI (in period two), oil prices, and gold prices, the COVID-19 cases, and interbank rates. In the third period, findings revealed that in the order of exchange rates, oil prices, the COVID-19 cases, gold prices, and interbank interest rates. The exchange rate order indicates in the fourth and the tenth periods are the COVID-19 cases, oil prices, gold prices, and interbank interest rates. In model 2 the shocks of JCI return determine the JCI return. Another stock index variable determining the JCI return in the second period is the stock indices of the United States, Thailand, Singapore, Japan, and Malaysia. In the third to fourth periods with Thailand, the United States, Singapore, Japan, and Malaysia. In the fifth period with the order of the stock indices of Thailand, the United States, Japan, Singapore, and Malaysia. In the sixth to tenth periods with the order of the stock indices of Thailand, the United States, Singapore, Japan, and Malaysia.

    Using both the impulse response function as depicted in Figures 1 and 2 and the variance decomposition analysis as depicted in Table 4. Our empirical strategy divulged that JCI return responds positively to its shock, the shock of world gold price, exchange rate, stock indices of Malaysia and Japan. While for the world oil price, interbank, COVID-19 cases, indices of Singapore, Thailand, and United States, findings from the impulse response function and variance decomposition analysis revealed both positive and negative responses of JCI return.

    5.3 Discussion

    The impulse response results indicate that the return of JCI responds negatively to the shocks of world oil prices. With this finding, it can be established that when there is an increase in world crude oil prices, there will be a decline in the returns of JCI. These results are in line with previous studies conducted before the COVID-19 period, which shows that changes in world oil prices negatively affect the country’s stock return (Giri and Joshi, 2017;Rahman and Mustafa, 2018;Khan et al., 2019;Darmawan et al., 2020). The negative relationship between the world crude oil price and stock return can occur due to exporter or importer of oil. The shock of world oil price which response negatively indicates that it is an importing country (Park and Ratti, 2008). The shock of world oil prices causes the manufacturing sector to experience higher costs and lower profits. Companies that have decreased profits experienced decreasing dividends distribution to investors (Masih et al., 2011). Rising oil prices pushed up inflation due to higher production and shipping costs. Inflation causes negative sentiment for the stock market (Youssef and Mokni, 2019).

    The result from impulse response indicates that the return of JCI responds positively to the shock of world gold price. It can be established that when the gold price increase, the return of JCI also increases. These results are in line with previous research before the emergence of the COVID-19, which shows that changes in gold price positively impact a country’s stock index (Nordin et al., 2014;Rahman and Mustafa, 2018;Robiyanto, 2018).

    Baur and Lucey (2010) show that the world gold price and the stock market are often negatively related. Gold is considered a safe haven when the stock market is in turmoil. Another opinion from Šoja (2019) shows that it is crucial to diversify investment instruments such as gold, stocks, and bonds simultaneously when there is economic disruption. A study by Mulyadi and Anwar (2012) indicates that a decrease in stock returns will be followed by an increase in gold returns. On the contrary, when gold returns increase, the stock return will also increase. The conclusion is, when an increase in gold prices occurs, stock instruments will respond positively.

    The result of impulse response indicates that JCI return responds positively to the shock of the rupiah exchange rate. It can be established that when there is an increase in the rupiah exchange rate, there will be an increase in the return of JCI. These results are in line with Giri and Joshi’s (2017) findings in the case of India’s Sensex index. As companies need other countries’ currencies for international transactions. The exchange rate will affect companies’ profits due to companies’ dependence on foreign raw materials (Šimáková, 2017). Depreciation of a country’s currency can cause an increase in its exports. The increase in exports is due to lower prices for exported goods in the destination countries. The increase in sales of goods also increased company profits, which led to positive sentiment for stock investors (Paramati and Gupta, 2013).

    The result of impulse response indicates that the return of JCI responds negatively to the shock of the interbank rate. It can be established that a rise in interest rates between IndONIA banks will cause a decrease in the return of JCI. These results align with Nordin et al.’s (2014) finding. The same results are reported by Robiyanto’s (2018) study. Interest rate is considered as an indicator of economic conditions. Interest rate is used to help understand the direction of the economy and the interest rate impact the stock market (Amarasinghe, 2015). Stimulatory interest rate policies lead to an increase in share prices. The market defines the comprehensive policy as an economic boost that positively impacts stock investors (Bissoon et al., 2016). Companies with capital loans for expansion can use lower interest rates. Companies with increased performance or profits can reflect the expectations of stock investors (Khan et al., 2017b).

    The result of impulse response indicates that the JCI return responds negatively to the additional daily COVID-19 cases in Indonesia. It can be concluded that when there is an increase in COVID-19 cases in Indonesia, the JCI return will decrease. This result is consistent with Al-Awadhi et al. (2020), who reported that the daily growth cases of COVID-19 and confirmed cases of death have harmed stock return contained in the Hang Seng and Shanghai indices. The COVID-19 virus, which is spreading rapidly, creates economic problems for affected countries (Khan et al., 2020). Social restrictions have increasingly hit the economic situation of countries affected by COVID-19. The policy of closing public places, economic activities, and offices has increasingly created uncertainty for market players (Ozili and Arun, 2020). News that informs about economic problems creates negative sentiment for the stock market (Tetlock, 2007).

    Finding from the impulse response indicates that the return of JCI responds positively to the shock of Malaysia and Japan stocks indices. It can be established that when there is an increase in the stocks indices of Malaysia and Japan, the return of JCI will rise. Different results are shown by other variables where the impulse response function shows that the return negatively responds to Singapore, Thailand, and the United States stock price index. With this finding, it can be established that when there is an increase in the stock indices of Singapore, Thailand, and the United States, there will be a decrease in the return of JCI. This result is in line with Octavia and Wijaya’s (2020) study. Previous studies have also revealed a correlation between the stock index of Southeast Asian countries and can be influenced by the stock indices of Japan and the United States (Thao and Daly, 2012;Lee and Chou, 2019;Wu, 2020). Technological advancement provides easy access for global investors to view and transact shares of other countries (Arouri and Foulquier, 2012). Integrating the stock market between countries causes a shift in the risk of increasing and decreasing share prices (Shen, 2018). The stock market of other countries can be used as a means of comparing the benefits of investing. Investment in other countries is a form of asset diversification from different political, economic, security, and market psychology among countries (Abbas et al., 2013).

    6. SOME CONCLUDING REMARKS

    In model 1, our empirical strategy from the impulse response function shows that the return of JCI has a positive response to the shock of the Malaysian and Japanese stock indices. Different results are obtained by other variables where the impulse response shows that the return has a negative response to the shocks of Singapore, Thailand, and the United States stock price indices. The JCI return is determined by the shock of itself. The shocks of other variables in determining the JCI return are exchange rates, oil prices, gold prices, the COVID-19 cases, and interbank interest rates. In model 2, the JCI return is determined by the shock of itself. Other variables that determine the JCI return are the United States, Thailand, Singapore, Japan, and Malaysia stock indices. Based on the study findings we offer the following policy recommendations;

    i. Since the JCI return responds negatively to the shock in oil price, the government needs to regulate the domestic oil price so that it is affordable. The government needs to provide subsidies when oil prices are high. Investors need to see the world oil market situation and the behaviour of other investors.

    ii. Since the JCI return responds positively to the shock in the gold price, the central bank needs to regulate or secure gold reserves as diversification of portfolio assets. Gold assets can be used as a central bank store of value when gold prices fall. Investors need to diversify into gold assets when the gold market situation is unstable.

    iii. Since the JCI return responds positively to the shock in the exchange rate, the central bank needs to regulate the exchange rate to ensure a stable economy. The central bank can use foreign reserves for market intervention when the exchange rate continues to weaken. Investors need to look at the situation of major world currencies such as the US dollar and the development of the United States central bank in issuing policies.

    iv. Since JCI return responds negatively to the shock in interbank rate, the central bank needs to regulate the interest rates using expansionary monetary policies. Investors need to see the market response to changes in daily interest rates.

    v. Since JCI return responds negatively to the shock in daily cases of COVID-19, the government needs to adopt a policy for market confidence and to provide relief through direct assistance. Investors need to see stock opportunities in specific sectors that are considered profitable when economic conditions are unstable.

    vi. Since JCI return has recorded mixed responses due to the shock in other countries’ stock indices, there is a need for trade rules in the Indonesia stock exchange to keep the JCI from high fluctuations in a short time. The government is expected to be able to socialize the benefits of investing in stocks. Domestic investors need to strengthen the Indonesian stock market to prevent it from any external shock.

    Limitation

    This study is not free from some limitations which range from the time and sample coverage as well as the limitations associated with the adopted analytical model. As the study focussed on the examination of Indonesia’s stock market returns to the internal and external shocks during COVID-19 associated with the Southeast Asia region (Malaysia, Singapore, and Thailand), East Asia region (Japan), and the United States. There is a need to explore more on stock market returns response by incorporating other regions’ shocks not included in this study. Additionally, the muddled limitation of the set model has been handled by applying VECM.

    Figure

    IEMS-21-1-85_F1.gif

    Analysis of impulse response: Model 1.

    IEMS-21-1-85_F2.gif

    Analysis of impulse response: Model 2.

    Table

    Stationarity test

    Lag test

    Test of cointegration

    Analysis of variance decomposition

    REFERENCES

    1. Abbas, Q. , Khan, S. , and Shah, S. Z. (2013), Volatility transmission in regional Asian stock markets, Emerging Markets Review, 16, 66-77.
    2. Aggarwal, R. and Lucey, B. M. (2005), Psychological barriers in gold prices?, Review of Financial Economics, 16(2), 217-230.
    3. Al-Awadhi, A. M. , Alsaifi, K. , Al-Awadhi, A. , and Alhammadi, S. (2020), Death and contagious infectious diseases: Impact of the COVID-19 virus on stock market returns, Journal of Behavioral and Experimental Finance, 27, 100326.
    4. Ali, M. and Sun, G. (2017), Dynamic relations between stock price and exchange rate: Evidence from South Asia, International Journal of Economics and Financial Issues, 7(3), 331-341.
    5. Ali, M. H. , Kareem, H. B. , and Barrak, J. I. (2020), Uses of the revised exchange rate to maintain fair value of assets under Covid-19 crisis, Industrial Engineering and Management Systems, 19(4), 932-948.
    6. Amarasinghe, A. (2015), Dynamic relationship between interest rate and stock price: Empirical evidence from Colombo stock exchange, International Journal of Business and Social Science, 6(4), 92-97.
    7. Apornak, A. (2021), Human resources allocation in the hospital emergency department during COVID-19 pandemic, International Journal of Healthcare Management, 14(1), 264-270.
    8. Arouri, M. E. H. and Foulquier, P. (2012), Financial market integration: Theory and empirical results, Economic Modelling, 29(2), 382-394.
    9. Baur, D. G. and Lucey, B. M. (2010), Is gold a hedge or a safe haven? An analysis of stocks, bonds and gold, The Financial Review, 45(2), 217-229.
    10. Baur, D. G. and McDermott, T. K. (2010), Is gold a safe haven? International evidence, Journal of Banking & Finance, 34(8), 1886-1898.
    11. Bissoon, R. Seetanah, B. , Babajee, R. B. , Ramdhany, N. G. , and Seetah, K. (2016), Monetary policy impact on stock return: Evidence from growing stock markets, Theoretical Economics Letters, 6(5), 1186-1195.
    12. Carollo, S. (2012), Understanding Oil Prices: A Guide To What Drives The Price of Oil in Today's Markets, John Wiley & Sons, Cornwall.
    13. Celebi, K. and Hönig, M. (2019), The impact of macroeconomic factors on the german stock market: Evidence for the crisis, pre- and post-crisis periods, International Journal of Financial Studies, 7(2), 1-13.
    14. Contuk, F. Y. , Burucu, H. , and Gungor, B. (2013), Effect of gold price volatility on stock returns: Example of Turkey, International Journal of Economics and Finance Studies, 5(1), 119-140.
    15. Darmawan, I. , Siregar, H. , Hakim, D. B. , and Manurung, A. H. (2020), The effect of crude oil price shocks on Indonesia stock market performance, Journal of Organization and Management, 16(1), 11-23.
    16. Demir, C. (2019), Macroeconomic determinants of stock market fluctuations: The case of BIST-100, Economies, 7(1), 1-14.
    17. Dong, X. and Yoon, S. M. (2019), What global economic factors drive emerging Asian stock market returns? Evidence from a dynamic model averaging approach, Economic Modelling, 77, 204-215.
    18. Enders, W. (2014), Applied Econometric Time Series, Wiley, Hoboken.
    19. Faghihi-Nezhad, M. and Minaei-Bidgoli, B. (2018), Prediction of stock market using an ensemble learning-based intelligent model, Industrial Engineering and Management Systems,17(3), 479-496.
    20. Fattouh, B. (2011), An Anatomy of the Crude Oil Pricing System, Oxford Institute for Energy Studies, Oxford.
    21. Giones, F. , Brem, A. , Pollack, J. M. , Michaelis, T. L. , Klyver, K. , and Brinckmann, J. (2020), Revising entrepreneurial action in response to exogenous shocks: Considering the COVID-19 pandemic, Journal of Business Venturing Insights, 14(2020), 1-7.
    22. Giri, A. K. and Joshi, P. (2017), The impact of macroeconomic indicators on Indian stock prices: An empirical analysis, Studies in Business and Economics, 12(1), 61-78.
    23. Gokmenoglu, K. K. and Fazlollahi, N. (2015), The interactions among gold, oil, and stock market: Evidence from S&P500, Procedia Economics and Finance, 25, 478-488.
    24. Goodell, J. W. (2020), COVID-19 and finance: Agendas for future research, Finance Research Letters, 35, 1-5.
    25. Gopinathan, R. and Durai, S. R. S. (2019), Stock market and macroeconomic variables: New evidence from India, Financial Innovation, 5(1), 1-17.
    26. Gujarati, D. N. and Porter, D. C. (2008), Basic Econometrics, McGraw-Hill, New York.
    27. Hussain, M. M. , Aamir, M. , Rasool, N. , Fayyaz, M. , and Mumtaz, M. (2012), The impact of macroeconomic variables on stock prices: An empirical analysis of Karachi stock exchange, Mediterranean Journal of Social Sciences, 3(3), 295-312.
    28. Jazairi, N. T. , Matatko, J. , Kotz, S. , Balakrishnan, N. , and Read, C. B. (2006), Encyclopedia of Statistical Sciences, John Wiley & Sons, New Jersey.
    29. Khan, J. and Khan, I. (2018), The impact of macroeconomic variables on stock prices: A case study of Karachi stock exchange, Journal of Economics and Sustainable Development, 9(13), 15-25.
    30. Khan, K. , Zhao, H. , Zhang, H. , and Yang, H. (2020), The impact of COVID-19 pandemic on stock markets: An empirical analysis of world major stock indices, Journal of Asian Finance, Economics and Business, 7(7), 463-474.
    31. Khan, M. K. (2019), Impact of exchange rate on stock returns in Shenzhen stock exchange: Analysis through ARDL approach, Akkaya Science Publishing, 1(2), 15-26.
    32. Khan, M. K. , Teng, J.-Z. , and Khan, M. I. (2019), Asymmetric impact of oil prices on stock returns in Shanghai stock exchange: Evidence from asymmetric ARDL model, Plos One, 14(4), 1-14.
    33. Khan, M. K. , Teng, J.-Z. , Parvaiz, J. , and Chaudhary, S. K. (2017a), Nexuses between economic factors and stock returns in China, International Journal of Economics and Finance, 9(9), 182-191.
    34. Khan, M. T. , Tan, S.-H. , Chong, L.-L. , and Ong, H.-B. (2017b), Investment characteristics, stock characteristics and portfolio diversification of finance pro-fessionals, Borsa Istanbul Review, 17(3), 164-177.
    35. Kwag, S. W. and Kim, Y. S. (2013), Stock price predictability of financial ratios and macroeconomic variables: A regulatory perspective, Industrial Engineering and Management Systems, 12(4), 406-415.
    36. Lawal, A. I. , Somoye, R. O. , Babajide, A. A. , and Nwanji, T. I. (2018), The effect of fiscal and monetary policies interaction on stock market performance: Evidence from Nigeria, Future Business Journal, 4(1), 16-33.
    37. Lee, C. and Chou, P. (2019), Structural breaks in the correlations between Asian and US stock markets, North American Journal of Economics and Finance, 51, 101087.
    38. Liu, H. Y. , Manzoor, A. , Wang, C. Y. , Zhang, L. , and Manzoor, Z. (2020), The COVID-19 outbreak and affected countries stock markets response, International Journal of Environmental Research and Public Health, 17(8), 1-19.
    39. Mankiw, N. G. (2016), Macroeconomics, Macmillan Worth Publishers, New York.
    40. Markwat, T. , Kole, E. , and Van Dijk, D. V. (2009), Contagion as a domino effect in global stock markets, Journal of Banking and Finance, 33(11), 1996-2012.
    41. Masih, R. , Peters, S. , and Mello, L. D. (2011), Oil price volatility and stock price fluctuations in an emerging market: Evidence from South Korea, Energy Economics, 33(5), 975-986.
    42. Mishkin, F. S. (2012), Macroeconomics Policy And Practice, Addison-Wesley Pearson Education, Boston.
    43. Mishra, M. K. (2020), The world after COVID-19 and its impact on global economy,ZBW - Leibniz Information Centre for Economics, Available from: https://www.econstor.eu/handle/10419/215931.
    44. Mulyadi, M. S. and Anwar, Y. (2012), Gold versus stock investment: An econometric analysis, International Journal of Development and Sustainability, 1(1), 1-7.
    45. Narang, S. and Singh, R. P. (2012), Causal relationship between gold price and Sensex: A study in Indian context, Vivekananda Journal of Research, 33-37, Available from: https://www.researchgate.net/publication/287196336_Causal_Relationship_between_Gold_Price_and_Sensex_A_Study_in_Indian_Context.
    46. Nordin, N. , Nordin, S. , and Ismail, R. (2014), The impact of commodity prices, interest rate and exchange rate on stock market performance: An empirical analysis from Malaysia, Malaysian Management Journal, 18, 39-52.
    47. Octavia, M. and Wijaya, C. (2020), Analysis of cointegration and causality relationship among selected stock market indexes in the world and Indonesia stock exchange composite index (IHSG) for the period 2005-2017, Open Journal of Business and Management, 8(3), 1226-1242.
    48. Aigbovo, O. and Izekor, A. O. (2015), The impact of macroeconomic variables on stock market index in Nigeria, African Journal of Management Sciences, 1(1), 18-40.
    49. Ozdurak, C. , Alcan, G. , and Guvenbas, S. D. (2020), The impact of Covid-19 to global pharmaceuticals and biotechnology company stocks returns, Journal of Business, Economics and Finance (JBEF), 9(2), 68-79.
    50. Ozili, P. K. and Arun, T. G. (2020), Spillover of COVID-19: Impact on the global economy, SSRN Electronic Journal. Available from: https://www.researchgate.net/publication/340236487_Spillover_of_COVID-19_impact_on_the_Global_Economy.
    51. Paramati, S. R. and Gupta, R. (2013), An Empirical relationship between exchange rates, interest rates and stock returns, European Journal of Economics, Finance and Administrative Sciences, 56, 168-181.
    52. Park, J. and Ratti, R. A. (2008), Oil price shocks and stock markets in the U.S. and 13 European countries, Energy Economics, 30(5), 2587-2608.
    53. Rahman, M. and Mustafa, M. (2018), Effects of crude oil and gold prices on us stock market: Evidence for USA from ARDL bounds testing, Finance and Market Universe Scientific Publishing, 3(1), 1-9.
    54. Ranjbar, M. H. , Sadeghi, N. , and Sadeghi, L. (2018), The effect of investment risk and macro-economy factors on stock return in D8 member states based on a dynamic panel approach, Industrial Engineering and Management Systems, 17(3), 613-622.
    55. Rauterberg, G. and Verstein, A. (2013), Index theory: The law, promise and failure of financial indices, Yale Journal on Regulation, 30(1), 1-62.
    56. Robiyanto, R. (2018), The effect of gold price changes, USD/IDR exchange rate changes and bank Indonesia (BI) rate on Jakarta composite index (JCI)’s return and Jakarta Islamic index (JII)’s return, Jurnal Manajaemen Dan Kewirausahaan, 20(1), 45-52.
    57. Sadono, Y. A. (2020), IDX implements provisions regarding trading halt for trading on the exchange, Available from: https://www.idx.co.id/: https://www.idx.co.id/berita/press-releasedetail/?emiten.
    58. Salisu, A. A. , Ebuh, G. U. , and Usman, N. (2020), Revisiting oil-stock nexus during COVID-19 pandemic: Some preliminary results, International Review of Economics and Finance, 69, 280-294.
    59. Salvatore, D. (2013), International Economics, Wiley, New York.
    60. Sbeiti, W. and Alshammari, T. (2010), Integration of stock markets in the GCC countries: An application of the ARDL bounds testing model, European Journal of Economics, Finance and Administrative Sciences, 20, 35-43.
    61. Shen, Y. (2018), International risk transmission of stock market movements, Economic Modelling, 69, 220-236.
    62. Šimáková, J. (2017), The impact of exchange rate movements on firm value in Visegrad countries, Acta Universitatis Agriculturae Et Silviculturae Mendelianae Brunensis, 65(6), 2105-2111.
    63. Singh, N. P. and Sharma, S. (2018), Cointegration and causality among dollar, oil, gold and Sensex across global financial crisis, Vision, 22(4), 365-376.
    64. Šoja, T. (2019), Gold in investment portfolio from perspective of European investor, The European Journal of Applied Economics, 16(1), 41-58.
    65. Tetlock, P. C. (2007), Giving content to investor sentiment: The role of media in the stock market, The Journal of Finance, 62(3), 1139-1168.
    66. Thao, T. P. and Daly, K. (2012), Long run relationship among the southeast Asian equity markets, Proceedings of the 2nd International Conference on Economics, Trade and Development, IPEDR, 36, IACSIT Press, Singapore.
    67. Topcu, M. and Gulal, O. S. (2020), The impact of COVID-19 on emerging stock markets, Finance Research Letters, 36, 101691.
    68. Tripathi, V. and Kumar, A. (2016), Long run relationship between aggregate stock prices and macroeconomic factors in BRICS stock markets, Emerging Trends in Management and Public Policy Regal Publications, 1-28.
    69. Tripathy, N. (2015), Stock market integration: Evidence from BRIC countries, International Journal of Business and Emerging Markets, 7(3), 286-306.
    70. Wahyudi, S. , Hersugondo, H. , Laksana, R. D. , and Rudy, R. (2017), Macroeconomic fundamental and stock price index in southeast Asia countries: A comparative study, International Journal of Economics and Financial Issues, 7(2), 182-187.
    71. Wang, W. and Park, H. (2021), How vulnerable are financial markets to COVID-19? A comparative study of the US and South Korea, Sustainability, 13(10), 1-18
    72. Wu, F. (2020), Stock market integration in east and southeast Asia: The role of global factors, International Review of Financial Analysis, 67, 101416.
    73. Youssef, M. and Mokni, K. (2019), Do crude oil prices drive the relationship between stock markets of oil-importing and oil-exporting countries?, Economies, 7(3), 70.
    Do not open for a day Close