Asian Journal of Economics, Finance and Management

Applying a consistent estimation technique on data robust empirical results which the private and public poly makers can rely on. This study examined the effects of foreign trade and Foreign Direct Investment on capital formation in Nigeria.  The objectives of the study were to analyze the effects of total export, total imports and foreign direct investment on capital formation over the period studied. The study sourced yearly time series data which spanned through 1980-2020 from the Central Bank of Nigeria Statistical Bulletin. The data was subjected to statistical analysis (the OLS and the robust least squares) using the Eviews 12.0. The post estimation result of the OLS showed a violation of the OLS assumption. Further check using leverage and influence plots revealed the presence of outliers. The study therefore employed the robust least square regression to overcome the shortcoming of the OLS regression. The information criteria showed that that the robust least square regression was superior to the OLS result, however, among Bi-square M-estimation, the Huber M-estimation, and MM-estimation class of robust regression, the MM-estimation was the best model with the least information criteria (Akaike info criterion = 59.66) the results showed that only total export was significant in affecting capital formation.