Abstract

This paper is an investigation of causal relationships that exist between macroeconomic variables in Nigeria context These variables are interest rate inflation rate exchange rate real gross domestic product and unemployment rate Often a variable can better be forecasted by introducing past and current values of some other variables in the ARMA model or its AR approximation We achieved this by employing an augmented VAR approach such as the procedure proposed by Toda-Yamamoto This current work included a unit-root test with trend break functions without a priori information Specifically we employed the extended Augmented Dickey-Fuller test through innovational outlier and additive outlier models The truncation parameter was selected using the t-sig and F-sig general to specific recursive techniques Unknown breakpoints were observed which indicates a strong connection with the data