Abstract

Modelling and forecasting of leather export in Ethiopia have been attempted by using vector auto-regression (VAR) and vector error correction (VEC) models, respectively. The variables are value of leather export, export price of leather, consumer price index (endogenous variables) and nominal exchange rate (exogenous variables), respectively. The series are seasonally adjusted through standard tests built in X-12 ARIMA program in E-Views 6 statistical software. Post seasonal adjustment tests also assured that all series were non-seasonal. Unit root tests of the series reveal that all the series are non-stationary at zero level and stationary after first difference. The result of Johansen test indicates the existence of two co-integration relation between the variables. This implies the legitimacy of vector error correction model (VEC) model of order one to the data. The final result shows that a VEC model of lag one with two co-integration equations best fits the data. Export price of leather has a negative effect on value of leather exports. A one percent increase in a unit price of leather export will cause 5.82% decrease in value of leather export in the long run. In the short run Exchange Rate has a negative effect on exports of leather as expected.

Author: M. K. Sharma and Getnet Memo

Received on: November, 2018

Accepted on: August, 2020