Long-Run and Short-Run Dynamics among the Sectoral Stock Indices: Empirical ResultsAs the condition of the same order of integration is met, the existence of long-run relationship between the sector indices is examined using the Engle-Granger two-step procedure and the Johansen and Juselius cointegration tests. The findings of the Engle-Granger test are reported in Table 4. According to the results, the null hypothesis of no cointegration is rejected at the 1% level of significance for all combinations of the ISE sector index pairs. This points out that the sector index prices are related in the long-run. Results of the Johansen and Juselius multivariate cointegration tests among six sector indices are given in Table 5.
Trace and maximum eigenvalue statistics show that there are six cointegrating relationships among sector indices. The statistical evidence supports the results obtained through the bivariate Engle and Granger cointegration analysis. The results are also in line with the findings reported by Arbelaez et al., Wang et al. and Patra and Poshakwale for various stock exchanges where sector indices have a tendency to move toward same direction at least in the long-run. This confirms that most of the sectors are more or less influenced by both the macroeconomic indicators and political events in the long-run. This statistically significant long-run relationship between six sector indices supports that there are no benefits from portfolio diversification in terms reduction in risk. In the presence of cointegrated variables, the Granger causality test requires VECM to capture both the short-run dynamics between time-series and their long-run equilibrium relationship. The results of the causality tests through ECM for sectoral indices are reported in Table 6. The lagged error correction terms are statistically significant, suggesting bidirectional causality in the long-run for all sectoral indices except Banking-Holding and Holding-Commerce pairs. Electronic Payday Loans Online

For the short-run dynamics, there is unidirectional causality running from Banking to Chemistry, Holding and Communication. This reveals that the Banking sector seems to be the most dominant and leading sector in the ISE. This strong influence can be explained by banking sector’s market capitalization value of 44 billion TL and its approximately 41% share of the total market capitalization of the ISE-100 index as of January 31, 2011. The Banking sector index includes 17 banks and they represent 30% of the total trading volume in the ISE in 20102. There is also bidirectional causality between Holding and Commerce whereas unidirectional causality from Chemistry to Commerce. Having a better knowledge of the nature of the sub-sector relationships is crucial for considerable number of market participants to make optimal portfolio allocation decisions across sectors. The noticeable correlation among sub-sector indices shows that the benefits of portfolio diversification diminish or may even completely disappear in ISE.

Table-4: The Results of Engle and Granger Cointegration Test

Index pairs ADF k
XBANK- XKMYA -46.994 0
XBANK – XHOLD -48.421 0
XBANK – XMANA -47.248 0
XBANK – XTCRT -21.710 5
XBANK – XILTM -29.030 2
XHOLD – XMANA -48.767 0
XHOLD – XKMYA -33.960 1
XHOLD – XTCRT -21.868 5
XHOLD – XILTM -33.025 1
XMANA – XKMYA -50.569 0
XMANA – XTCRT -49.046 0
XMANA – XILTM -48.890 0
XKMYA – XTCRT -51.126 0
XKMYA – XILTM -21.332 5
XTCRT – XILTM -21.952 5

Table-5: The Results of Johansen-Juselius Maximum Likelihood Cointegration Tests

Maximum eigenvalue test Trace test
Null Alternative Statistic 95%criticalvalue Null Alternative Statistic 95%criticalvalue
r = 0 r = 1 537.15 43.41 r = 0 r > 1 853.28 107.34
r < 1 r = 2 509.88 37.16 r < 1 r > 2 816.12 79.34
r < 2 r = 3 485.85 30.81 r < 2 r > 3 806.24 55.24
r < 3 r = 4 469.96 24.25 r < 3 r > 4 520.39 35.01
r < 4 r = 5 434.34 17.14 r < 4 r > 5 850.42 18.39
r < 5 r = 6 416.08 3.84 r < 5 r > 6 416.08 3.84

Table-6: The results of Granger Causality through ECM

Index Pairs F-statistics Causal Inference EC (t-1)
1 XBANK – XKMYA 2. 328 X causes Y 0.624 (27.047)
XKMYA – XBANK 0. 007 No causality 0.253 (8.268)
2 XBANK – XHOLD 2.861 X causes Y 0.042 (0.639
XHOLD – XBANK 1.236 No causality -0.934 (-13.258)
3 XBANK – XMANA 1.622 No causality 0.496 (31.020)
XMANA – XBANK 0.013 No causality 0.231 (12.550)
4 XBANK – XTCRT 0.009 No causality 0.297 (5.073)
XTCRT-XBANK 2.023 No causality -0.702 (-9.727)
5 XBANK – XILTM 2.854 X causes Y -0.287 (-24.833)
XILTM – XBANK 0.916 No causality -0.198 (-18.259)
6 XHOLD – XMANA 0.372 No causality 0.825 (26.227)
XMANA – XHOLD 0.147 No causality 0.254 (7.855)
7 XHOLD – XKMYA 1.549 No causality 0.833 (14.240)
XKMYA – XHOLD 1.136 No causality 0.153 (2.297)
8 XHOLD – XTCRT 3.025 Bidirectional 0.723 (12.886)
XTCRT – XHOLD 2.716 Bidirectional -0.099 (-1.505)
9 XHOLD – XILTM 0.893 No causality 0.176 (29.901)
XILTM – XHOLD 0.152 No causality 0.082 (15.395)
10 XMANA – XKMYA 1.213 No causality 0.586 (28.927)
XKMYA – XMANA 0.603 No causality 0.247 (9.151)
11 XMANA – XTCRT 0.312 No causality -0.232 (-6.961)
XTCRT – XMANA 1.671 No causality -1.096 (-30.881)
12 XMANA – XILTM 0.912 No causality -0.189 (-35.890)
XILTM – XMANA 0.783 No causality -0.116 (-22.478)
13 XKMYA – XTCRT 2.693 X causes Y -0.260 (-4.889)
XTCRT – XKMYA 1.264 No causality -1.214 (-24.092)
14 XKMYA – XILTM 0.976 No causality -0.536 (-33.982)
XILTM – XKMYA 0.389 No causality -0.347 (-28.088)
15 XTCRT – XILTM 1.109 No causality 0.447 (28.264)
XILTM – XTCRT 0.799 No causality 0.090 (7.048)