Jurnal ISEI Tahun 2009
ANALISIS KESEIMBANGAN DAN HUBUNGAN SIMULTAN ANTARA VARIABEL EKONOMI MAKRO TERHADAP INDEKS HARGA SAHAM DI BURSA EFEK JAKARTA DENGAN METODE VAR ( VECTOR AUTOREGRESSION ) DAN ECM ( ERROR CORRECTION MODEL *)
This research of this study is to conduct empirical test the long term equilibrium and simultaneous relationship between macroeconomic variable stock price index at Jakarta Stock Exchange. For this purpose this study employs co integration model, vector auto regression ( VAR ), and Error Correction Model ( ECM ). The variable of this study consist of Composite Stock Price Index ( IHSG ), Interest Rate on Time Deposit ( one month and twelve month ) ( Depo 1 and Depo 12 ), SBI?s Discount Rate ( SBI ), Money Supply ( M1 and M2 ), Exchange Rate Rupiah to US dollar ( Rupiah ), and Inflation ( Inflasi ), of January 1990 to December 2000 period.
The result of this study by using graphic mode, autocorrelation Function ( ACF ), as well as unit roots by using Augment Dickey Fuller ( ADF ) method and Philips ? Perron ( PP ) method indicate that original data is no stationary, and stationary 1st degree difference. For the purpose of this research, after conducting causality test can be concluded that only four variables that have causality relationship the are : IHSG , Depo 1 , SBI, and Rupiah. Empirical test cointegration1990 : 01 to 2000 : 12 period some capable showing long term equilibrium compared with co integration test 1990 : 01 to 1997 : 07 period ( the period before monetary crisis ). In this respect this study is not able to indicate long term empirical equilibrium before and after monetary crisis in Indonesia. From the graphic normalized vector co integration 1990 : 01 to 2000 : 12 period also indicated that along period before crisis normalize graphic is enough stationary compared to monetary crisis period and this study is supported by eigen value which is decrease at the end of 1997 until in the middle of 1998. The co integration test also indicated that empirically there are non monetary events and extraordinary events ( for example : political events ) which is influence significantly to variables under study. So does with VAR method and ECM method 1990 : 01 to 2000 : 12 period which is show impulse response to one S. D. innovation and variance decomposition and proved empirically that Rupiah variable is more capable in explain its influence to IHSG, Depo 1 , and SBI?s variable in the long term and short term ( adjustment ).
To know relationship and validity of prediction, in this study is compared with other prediction method such as : Ordinary Least Square ( OLS bivariate ), and Mean Absolute Percent Error ( MAPE ). The finding of this study indicated that empirically VAR and ECM model more capable in explaining if among research variables there is causality relationship and simultaneous relationship.
Keywords : macroeconomic variable, Stock Price Index, vector auto regression, Error