A study on prediction model estimation of financial assets (exchange rate, KOSPI index, interest rate)

Abstract

In this study, financial assets such as exchange rate, KOSPI index, and interest rate (3-year government bond) were predicted using the vector error correction model used in various financial markets. For this purpose, time series data from February 2000 to January 2021 provided by the Bank of Korea were used. To estimate the prediction model, the stability of the time series variables was confirmed by the ADF test, the causal relationship between the time series variables by the Granger causality test, and the cointegration relationship between the time series variables by the cointegration test. In addition, the prediction model was estimated by identifying the model based on the minimum information criterion based on AICC statistics, and the validity of the model was confirmed by the significance test of the cross-correlation matrix and the multivariate Portmanteau test. As the result of forecasting with the estimated prediction model, the exchange rate was predicted to rise steadily, the KOSPI index was predicted to fall, but remain in the mid-3,000 range, and the interest rate (3-year government bond) was predicted to decline.