1. E-mail: eavci@marmara.edu.tr E-mail: belagha@ucas.edu.ps
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market
Bassam Al-Agha
University College of Applied Sciences, Gaza, Palestine
Assoc. Prof. Dr. Emin Avci
Department of Business Administration, Faculty of Business Administration, Marmara University, Istanbul, Turkey
Bassam Al-Agha
Department of Administration and Financial Sciences, University College of Applied Sciences, Gaza, Palestine

2. Outline Introduction Object of the study Important of study
Literature review
Data and methodology
Analysis
Results
Conclusion
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

3. Introduction
It is recognized that global stock markets relationships play an important role in the decisions of the international investors all over the world.
The correlations among the stock markets become an object of an interest for the policymakers because they indicate for the stability of the global financial system.
To design a good portfolio, the investors should understand how the stock markets returns are related each to other.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

4. Introduction
Islamic stock market becomes an important part of the international stock markets; it has now evolved to become an integral part of the international financial system.
While the process of the international financial integration gets more intensified, the question, what is the degree of co-movement between the Islamic stock markets and their conventional counterparts, still remains open.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

5. The object of study
The purpose of this study is to examine the short- term and the long-term co-movement relationship among conventional stock markets and Islamic stock markets.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

6. Research Question
What is the degree of co-movement relationship between the Islamic stock market and conventional stock market?
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

7. Significant of study
This study is significant because it will help decision-makers, especially in the case of diversification.
Will give more knowledge about Islamic and conventional stock market performance.
And as regards for Islamic performance studies, limited studies had been found.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

8. Literature Review:
The previous studies, which deal with the relationship and co-movement between financial stock markets, can be classified into three categories namely: studies that investigates and compares the relationships between Islamic stock markets and conventional stock markets, the second one focus on the linkage between the Islamic markets across countries, and the last category addresses the relationships between the conventional markets among themselves.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

9. Previous studies Park & Masih (2015)
Study
Findings
Park & Masih (2015)
the co-integration between the DJIM index and three conventional indexes such as DAX, Hang Seng, and KL is not exist at all. Nonetheless, S&P and DJIM are co-integrated, and that co-integration explains the long-term theoretical relations amongst the indexes.
Saiti & Masih (2014)
there is a theoretical long-run co-movement in the Shariah China Index with the Islamic and conventional markets that the Shariah China Index is Granger-caused by all the conventional and Islamic markets.
Dewandaru et al., (2014)
weakness in the stock market integration amongst the equity markets of GCC equity and other international Islamic indices.
Achsani et al., (2007)
whereas there are significant correlations between the Islamic stock indices of US and Canada, Indonesia and Malaysia, and Japan and Asia Pacific. On the other hand, USA Islamic stock market has strong effect on the other Islamic stock markets.
Conversely, the Canadian, Malaysian, Indonesian and Asia Pacific stock markets have a weak influence on the Islamic stock market of USA.

10. Previous studies Hussin et al., (2013)
Study
Findings
Hussin et al., (2013)
absence of the integration between the Islamic stock market of Malaysia and Indonesia as well as the world markets.
Albaity & Mudor (2012)
results of the study which is conducted in the USA and Malaysia showed the absence of difference in returns between the Islamic indices and conventional ones.
Kassim & Kamil (2012)
there is no co-integration among the Islamic stock markets in both pre and during crisis. Therefore, the long-run co-movements among the Islamic stock markets were immune to the 2007 crisis.
Kassim (2013)
The results of the research point that the integration level of the Islamic stock market was exposed to many changes due to the global financial crisis. The level of development and the geographical factors have affected on the Islamic stock markets integration.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

11. Data and Methodology
Five countries are selected based on the availability of both the conventional indices and Islamic indices (Table 1).
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

12. Table1: Selected indices for the study
Country
Name of the index
Symbol
Type
USA
Dow Jones Islamic Market U.S. Index
DJIMUS
Islamic
S&P 500 Composite Index
S&PCOMP
Conventional UK
Dow Jones Islamic Market U.K. Index
DJIMUK
FTSE 100 Index
FTSE100
Canada
Dow Jones Islamic Market Canada Index
DJIMCN
 S&P/TSX Composite Index
S&P/TSX
Japan
Dow Jones Islamic Market Japan Index
DJIMJAP
Nikkei 225 stock average index
JAPDOWA
Indonesia
Jakarta Islamic Index
JAKISLM
Jakarta Composite Index
JAKCOMP
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

13. The co-movement between conventional stock index and Islamic stock indices
Unit Root test
The results of ADF are shown in Table 2. The null hypothesis of a unit root is not rejected for all indices in log level, whereas it is rejected when it is taken in their log first differences.
So the results support the stationary of series at level, then we can say that it satisfy the necessary conditions for co-integration test.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

14. Data and Methodology
Applying tests to monthly data from January to December 2014
Methodology of this study relies on the Vector Autoregression (VAR), Granger Causality test and Engle- Granger co-integration test.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

15. Analysis
Unit root: before appling of time series analysis, the time series to be analyzed should be stationary. For this, Augmented Dickey-Fuller (ADF) unit root test was utilized. This is very important to avoid spurious regression.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

16. Augmented Dickey-Fuller test- ADF
Table2: Unit Root test Results for series of study
Country
Variables
Augmented Dickey-Fuller test- ADF
Level
1st Diff.
Statistic
P-value
Canada
Log(DJIMCN)
-1.97
0.2992
-13.29
0.0000
Log(S&P/TSX)
-1.11
0.7104
-11.29 UK
Log(DJIMUK)
-1.78
0.3904
-15.27
Log(FTSE100)
-1.89
0.3348
-13.64
USA
Log(DJIMUS)
-0.65
0.8550
-13.16
Log(S&PCOMP)
-0.79
0.8200
-12.52
Japan
Log(DJIMJAP)
-3.17
0.0233
-12.08
Log(JAPDOWA)
-2.13
0.2328
-11.45
Indonesia
Log(JAKISLM)
-0.90
0.7857
-10.68
Log(JAKCOMP)
-0.35
0.9138
-10.20
Average
Log (Islamic)
-1.47
0.5448
-12.82
Log (conventional)
-1.36
0.6026
-10.72
Note: All monthly returns were calculated as log using monthly closing prices.

17. VAR
Prior to carrying out the co-integration test, we applied VAR analysis to know the number of the lags needed for co-integration test.
Table 3 shows the results of the selected length of lags for each VAR model in each country according to Final Prediction Error (FPE) and Akaike Information Criteria (AIC). .
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

18. VAR country Canada UK USA Japan Indonesia Average criteria FPE AIC
Optimal length 6 2 3
Table 3: Optimal length of lags
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

19. VAR Lag Exclusion Wald Test
Wald test is used to examine the lags that should be excluded from the analysis. According to the Wald test if the p-value is greater than 0.05 level then this lag must be excluded and vice versa.
Table 4: Lag Exclusion Wald Tests results
Canada UK
USA
DJIMCN
S&P/TSX
DJIMUK
FTSE100
DJIMUS
S&PCOMP
Lag Excluded
2,4,5
2,3,4,5 2
1,2
Japan
Indonesia
Average
DJIMJAP
JAPDOWA
JAKISLM
JAKCOMP
Islamic
Conventional
2,3

20. VAR Model Estimation
Table 5 shows the relationship between Islamic and conventional indices for each country and the average of all countries.
So the results of VAR model as follow: For Canada and Indonesia, we notice that there is a two-way relationship, which means that Islamic and conventional stock markets affect each other, while the effect of conventional stock markets on Islamic index is stronger.
For UK, USA and Japan there is only a one-way relationship, which means that only conventional stock markets affect the Islamic stock markets. Regarding to the average of the stock markets, a two-way relationship is found between the Islamic and conventional markets, which means that the two kinds of stock markets affect each other. Moreover, we notice that this effect is approximately equal.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

21. VAR Model Estimation Canada UK USA lags DJIMCN S&P/TSX DJIMUK FTSE100
DJIMUS
S&PCOMP
DJIMCN(-1)
-0.572
-0.218*
DJIMUK(-1)
-0.441 -
DJIMUS(-2)
0.499
DJIMCN(-3)
0.333 --
FTSE100(-1)
0.389*
S&PCOMP(-2)
-0.589*
DJIMCN(-6)
0.652
0.385*
S&P/TSX(-1)
1.113*
0.513
S&P/TSX(-3)
-0.697*
S&P/TSX(-6)
-1.045*
-0.685
Constant
0.000
0.003
0.002
F-Statistic
6.565
7.176
4.925
2.966
R-square
0.192
0.146
0.053
0.033
Japan
Indonesia
Average
DJIMJAP
JAPDOWA
JAKISLM
JAKCOMP
Islamic
Conventional
DJIMJAP(-1)
-0.161
JAKISLM(-1)
0.213
0.261*
Islamic(-1)
-0.578
-0.436*
JAPDOWA(-1)
0.264*
JAKCOMP(-1)
0.312*
0.257
Conventional(-1)
0.804*
0.654
Islamic(-3)
0.222*
Conventional(-3)
-0.110
-0.002
0.006
0.005
-0.001
4.029
12.597
14.949
9.183
6.023
0.044
0.126
0.095
0.124

22. Granger Causality
In order to examine the direction of the relation between indexes, Granger Causality is tested at the first three lags (1, 2, and 3) for each country.
Table 6 shows the results of Granger Causality; it can be note that there is a Granger Causality in two ways in Canada, which means Islamic and conventional stock markets affect each other. For UK, USA and Japan also there is only one-way Granger Causality toward Islamic markets, which means that conventional stock markets cause Islamic stock markets. For Indonesia, no Granger Causality exists between Islamic and conventional stock markets. Regarding the average of the stock markets, a two-way Granger Causality exists which means Islamic and conventional stock markets affect each other. These results significantly match with the results of the previous VAR analysis except for Indonesia.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

23. Table 6: Granger Causality test results.
Country
Granger Causality
Lag = 1
Lag = 2
Lag = 3 F
P-Value
Canada
S&P/TSX does not Granger Cause DJIMCN
12.61
0.0005
6.57
0.0018
6.87
0.0002
DJIMCN does not Granger Cause S&P/TSX
5.39
0.0214
2.60
0.0769
2.88
0.0375 UK
DJIMUK does not Granger Cause FTSE100
0.95
0.3300
1.08
0.3409
0.4170
FTSE100 does not Granger Cause DJIMUK
6.42
0.0122
3.07
0.0489
2.22
0.0881
USA
DJIMUS does not Granger Cause S&PCOMP
0.03
0.8588
1.66
0.1926
1.20
0.3097
S&PCOMP does not Granger Cause DJIMUS
0.61
0.4374
3.05
0.0499
Japan
DJIMJAP does not Granger Cause JAPDOWA
0.45
0.5029
0.20
0.8161
0.44
0.7245
JAPDOWA does not Granger Cause DJIMJAP
6.16
0.0140
2.21
0.1133
2.06
0.1071
Indonesia
JAKISLM does not Granger Cause JAKCOMP
2.05
0.1539
1.68
0.1900
1.44
0.2329
JAKCOMP does not Granger Cause JAKISLM
0.90
0.3450
0.22
0.8066
0.43
0.7342
Average
Islamic does not Granger Cause Conventional
9.90
0.0019
5.79
0.0037
4.76
0.0033
Conventional does not Granger Cause Islamic
18.13
0.0000
10.58
7.70

24. Engle- Granger Co-integration Test
The co-integration test is one of the important tests, in studying the linkage among stock market, that it examine whether the indices are theoretically linked or not. If the indices are co-integrated each other, it means that there is a co-movement among these stock markets in the long term reaching the equilibrium, even they maybe move differently in the short term (Park & Masih, 2015).
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

25. Engle- Granger Co-integration Test
Table 7 shows the results of co-integration between closing share price series for Islamic indices and conventional indices in each country using Engle- Granger co-integration test. The results indicate that all of the test P-values are greater than So we can accept the null hypotheses “Ho: series are not co-integrated”. This means that there is no co-integration between series in each country in long term. According to this result, the investors can utilize the portfolio diversification between the Islamic and conventional stock market to reduce the risk in long term.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

26. Table 7 Engle- Granger co-integration test results.
Country
Dependent
z-statistic
P-Value.*
Canada
Log(DJIMCN)
-9.28
0.3994
Log(S&P/TSX)
-5.02
0.7282 UK
Log(DJIMUK)
-4.69
0.7540
Log(FTSE100)
-11.26
0.2838
USA
Log(DJIMUS)
-4.61
0.7613
Log(S&PCOMP)
-5.29
0.7057
Japan
Log(DJIMJAP)
-11.70
0.2621
Log(JAPDOWA)
-7.16
0.5534
Indonesia
Log(JAKISLM)
-16.63
0.1002
Log(JAKCOMP)
-15.73
0.1206
Average
Log (Islamic)
-7.53
0.5242
Log (Conventional)
-5.84
0.6602
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

27. Conclusion
This paper attempts to investigate the co-movement: between five Islamic indices and five conventional indices in UAS, UK, Canada, Japan and Indonesia.
The results are as the following, in short-term, Canadian and Indonesian Islamic and conventional stock markets affect each other. In UK, USA and Japan, there is only a one-way relationship, which means that only conventional stock markets affect the Islamic stock markets. But for average of the stock markets, two types of stock markets affect by each other in short-term.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

28. Conclusion
The result from Engle- Granger Co-integration Test pointed out that, there is no co-integration between Islamic indices and conventional indices in long-term.
In other words, it can be concluded the finding of this study showed that, there is a co-movement relationship between Islamic stock markets and conventional stock market in short-term, but this relationship disappears in long-term.
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha

29. Questions & Answerers
The Co-Movement Relationship Between Islamic Stock Market and Conventional Stock Market Avci and Al-Agha