Dynamic Relationship between Production Growth Rates of Three Major Cereals in Ghana

Abstract

Cereals play a major role in contributing to agricultural Gross Domestic Product and the economy, and are also used for the preparation of several local dishes and drinks in Ghana. Several linkages have been hypothesised of the relationship among the production growth rates of cereals. This study employed a Vector Autoregressive (VAR) model to investigate the relationship between the production growth rates of three major cereals in Ghana. The VAR model favoured VAR at lag 1 which indicated that, in addition to the bivariate unidirectional production growth rate causalities; there is also a bilateral causality between production growth rate in Millet and production growth rate in Milled Rice and a Rice to Corn unidirectional production growth rate causality. A diagnostic test revealed that the VAR (1) model was stable as it satisfies the stability condition. Also, the univariate ARCH-LM test and Ljung-Box test revealed that the model is free from conditional heteroscedasticity and serial correlation respectively. The Impulse Response Function and the Forecast Error Variance Decomposition were further employed to interpret the VAR (1) model. The Forecast Error Variance Decomposition revealed that growth rate in Millet production explains an appreciable amount of the forecast uncertainty in Rice and Corn.

Keywords: Production, Growth rates, Corn, Millet, Milled Rice, Granger-causality.