An econometric analysis of ISE market efficiency weak and semi strong forms

This study aims to investigate whether or not there are the co-integration and the causality relationships between stock prices and macroeconomic variables (Real Gross National Product, Consumer Price Index, Wholesale Price Index, Ml, M2 and M3) for the period of 1987:01-2002:12 and to test weak and semi-stronged (informationally) efficiency of ISE Market (Istanbul Stock Exchange). In Study, Augmented Dickey Fuller (ADF). Perron unit root tests are being employed for stationary of variables. The aspects of causality and the existing of causality relations have been determined by using Granger causality test and error correction model. Long term co-integration relationships between İMKB-100 and macroeconomic variables have been determined by Engle-Granger and Johansen-Juselius co-integration tests. According to ADF and Perron unit root tests, ISE index isn't stationary with models of constant and constant-trend in the level. The results of co-integration tests show that there is co-integration between ISE index and macroeconomics variables except for TUFE that is stationary in second level. According to the results of causality tests, there is bilateral causality relationship between ISE index and M3 money supply. While ISE index Granger causes real exchange rate and real gross national product, there is a causality relationship from TEFE to ISE index. The results show that ISE 1 00 index follows a random walk and ISE has a weak form efficiency. Because of that ISE 100 index can be estimated by using co-integration and causality relationships between macroeconomic variables and ISE 100 index, some investigators may have higher profit than the others. In addition, the results of causality and co-integration tests indicate that ISE market has no a semi-strong efficiency. VII