ϰϰ
rate in first quarter 2010 at 6,5 and experienced its lowest rate in second quarter 2012 until second quarter 2013 at 5.75 and the highest rate in the
last quarter of 2014 such amount 7,5. In term of JII it experienced relatively positive sentiment during
2010 until 2014. There are some negative sentiment in Quarter 4 2011, Quarter 3 2012, Quarter 1 2013, Quarter 3 Quarter 4 2013 and Quarter 1
4 2014. It is influenced by the general trend in Indonesia and International capital market after the US financial crisis in 2008 and 2012.
4.2 Research Methodology
This research methodology applies these following econometric procedures to analyze the relationship between stock returns and
macroeconomic variables: a. Descriptive Statistics
b. Unit Root Testing c. Cointegration Test
d. Vector Error Correction Model Estimation e. Variance Decomposition Test
4.3 Descriptive Statistics
Descriptive statistics represents the entire population or sample in the study and explains data central tendency and dispersion measurement.
Eviews 8 is used as the econometric tool to analyze the data. Eventually, this study finds the following descriptive statistics:
ϰϱ
Table 4.2 Descriptive Statistic
JII DJIM FHSI LNEXR ATE
CPI LNM2 BIR Mean
572.37 2339.27
11973.00 9.190
127.32 14.90
6.57
Median 574.40
2306.17 11904.81
9.151 126.92
14.93 6.5
Maximum 691.13
2935.41 14942.25
9.418 146.84
15.24 7.75
Minimum 413.73
1765.25 9049.27
9.046 113.78
14.54 5.75
Std. Dev.
74.20 309.81
1804.60 0.118
9.770 0.207
0.672
Skewness
-0.132 0.301
0.107 0.789
0.291 -0.173
0.209
Observation
60 60
60 60 60
60 60
Table 4.2 defines the descriptive statistics of all variable with regards to changes JII stock returns, DJIM stock returns, FHSI stock
returns and macroeconomic variables. It defines that first; Stock Returns of Jakarta Islamic Index JII Mean, Median, Maximum and Minimum value
are 572.37, 574.40, 691.13 and 413.73 respectively. Second, Dow Jones Islamic Market Index DJIM Mean, Median, Maximum and Minimum
value are 2339.27, 2306.17, 2935.41 and 1765.25 respectively. Third, FTSE Bursa Malaysia Hijrah Shariah Index FHSI Mean, Median,
Maximum and Minimum value are
11973.00
, 11904.81, 14942.25 and 9049.27 respectively. Forth, Exchange Rate between US Dollar to Rupiah
LNEXRATE Mean, Median, Maximum and Minimum value are 9.190, 9.151, 9.418 and 9.046 respectively. Fifth, Consumer Price Index CPI
Mean, Median, Maximum and Minimum value are 127.32, 126.92, 146.84 and 110.99 respectively. Sixth, Money Supply LNM2 Mean, Median,
Maximum adn Minimum value are 14.904, 14.931, 15.241 and 14.541 respectively. Seventh, Bank Indonesia Interest Rate BIR Mean, Median,
Maximum and Minimum value are 6.57, 6.50, 7.75 and 5.75 respectively.
ϰϳ
of the seven variables in this study in level and first difference. The unit root result from each variables are shown in the table below.
Table 4.3 Unit Root Test
± Augmented Dickey-Fuller ADF
Test ADF Variables Level Prob Note
First Difference
Pro b
Note Trend
Intercept Trend
Intercept JII
-3.021409 0.1462 Not
Stationary -8.287463 0.0
Stationary
DJIM -2.822243 0.1954
Not Stationary
-8.417526 0.0 Stationary
FHSI -3.043627 0.1439
Not Stationary
-8.738208 0.0 Stationary
LNEXRA TE
-3.102472 0.1153 Not
Stationary -7.525673 0.0
Stationary
CPI -1.476206 0.8264
Not Stationary
-7.237206 0.0 Stationary
LNM2 -2.842525 0.1886
Not Stationary
-9.723508 0.0 Stationary
BIR -1.426597 0.8426
Not Stationary
-4.343950 0.00 54
Stationary Note :
Lag length: 2 based on Lag length determination test McKinnon critics value on
Į 1 McKinnon critics value on
Į 5 McKinnon critics value on
Į 10
The result in the table above show that Jakarta Islamic Index JII, Dow Jones Islamic Market Index DJIM, FTSE Bursa Malaysia Hijrah
Shariah Index FHSI, exchange rate LNEXRATE, Consumer Price Index CPI, money supply LNM2 and Bank Indonesia Interest Rate
ϰϵ
Table 4.4 Lag Length Criteria
Table 4.4 shows the result of automatic lag length determination by Eviews 8. It shows that based on the value ofSchwarz information criterion
SC and Hannan-Quinn InformationHQ match in lag 1 which are 28.78130 and 27.52783.. The value of LR statistic and Final prediction
error FPE match in lag 4 which are 82.72074 and 447.5748 respectively. Meanwhile, Akaike information criterion AIC match in lag 5 which is
22.22053 respectively.Automatic Eviews7 lag length determines lag 4 as proper lag length in VECM estimation.
4.5 Cointegration Test
Cointegration test conducted by examine the stationarity from each variable in this research. Those variables must be integrated at order or the
similar degree. The integrating variables define similar stochastic trend and long-term similar movement. Cointegration test examined by the Johansen
Lag LogL
LR FPE
AIC SC
HQ
-1017.933 NA
36221471 37.27027
37.52575 37.36907
1 -679.2804
578.7874 979.3353
26.73747 28.78130
27.52783 2
-614.0852 94.82937
591.4283 26.14855
29.98073 27.63049
3 -575.8915
45.83243 1104.838
26.54151 32.16204
28.71501 4
-488.3984 82.72074
447.5748 25.14176
32.55064 28.00683
5 -414.0645
51.35795 471.8662
24.22053 33.41776
27.77717 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
ϱϬ
Juselius Cointegration test. In order to find out the number of cointegrating vectors, Trace statistic and Maximal Eigen value tests were used. The
UHVXOWRIWKH-RKDQVHQ¶V7UDFHDQG0D[LJHQYDOXHWHVWDUHVKRZQRQ7DEOH 4.5 below.
Table 4.5 Cointegration Test
± Johansen Juselius
Model Hypothe sized
No. Of CEs
Statist ical
Trace Critic
al Value
5 Maximu
m Eigen Statistic
al Trace Critic
al Value
5 Variab
le Long-
term Coefficien
t Elasticity
Result s
Lag Length
= 4
None 281.25 139.27 112.09 49.58
JII 1.00000 Trace
and Max
Eigen statisti
cs showe
d there are 4
cointeg ration
vectors At most
1 169.15
107.34 55.06 43.41 DJIM 0.408 At most
2 114.09
79.34 45.92 37.16 FHSI -0.202 At most
3 68.17 55.24 29.59 30.81 LNEX
RATE 1103.99
At most 4
38.58 35.01 22.48 24.25 CPI 1.216
At most 5
16.09 18.39 15.55 17.14 LNM2 8252.34 At most
6 0.54 3.84 0.54 3.84 BIR 113.82
: Denote significance at 5 respectively : MacKinnon-Haug-Michelis 1999 p-values
: Lag length based on Akaike Information Criterion AIC
Table 4.5 display the result of Johansen ± Juselius Cointegration
Test based on the likelihood ratio test consists of first trace statistics and
ϱϭ
second maximum Eigen value statistics. The results of trace statistics and maximum Eigen value were at 2 months lag length. The MacKinnon-
Haug-Michelis p- YDOXHVDUHDQGĮ 7KHUHIRUHZH
reject Ho and Accept Ha or the model is significant. Trace test and Max Eigen statistics showed the existence of 5 cointegrating vectors among
YDULDEOHVDWWKHĮ ,QWKHRWKHUZRUGWKHUHLV cointegration among JII, DJIM, FHSI, LNEXRATE, CPI, LNM2 and BIR for periods 2010 to
2014. Based on Johansen and Juselius Cointegration test, the first
normalized co-integrated vector towards JII variable using lag period proposed by AIC indicate long term relationship among DJIM, FHSI,
LNEXRATE, CPI, LNM2, BIR and stock returns of Jakarta Islamic Index JII. The result of the cointegration relationship can be summarized in
table 4.5 above. These value represent long-term elasticity measures, due to logarithmic transformation of JII, FHSI, CPI, M2 and BIR. Thus the
cointegration relationship can be re-expressed in table 4.6 below.
Table 4.6 Long-term Relationship
Dependent variable JII
Independent Variables
DJIM FHSI LNEXRATE CPI
LNM2 BIR
Coefficient 0.408
-0.202 1103.99
1.216 8252.34
113.82 t Value
4.743 6.929
5.531 0.220
12.728 5.278
, , Denote significance at 1, 5 and 10 respectively
The first normalized equation was estimated as below:
RJII =0.408 RDJIM - 0.202 FHSI + 1103.99 LNEXRATE + 1.216 CPI + 8252.34 LNM2 + 113.82 BIR