An Investigation of Agricultural Farm Dynamics using Bivariate Binary Regression

Abstract

Bivariate binary regression is a convenient way to identify the covariates associated with two correlated binary variables. At the same time, logarithm of odds ratio can be utilized to cross-check their interactions validity. Model parameters can be estimated in Bayesian approach with available prior information. This paper uses data obtained from a longitudinal survey conducted in 2002–2005. The aim of this survey was to make a guideline for the agricultural development and food security in Africa. Using available variables, the terms ‘Extensification’ and ‘Intensification’ was defined as two traditions for the farm dynamics. A simple data analysis is done using bivariate binary regression in WinBUGS. Results indicate that some factors, for instance, ‘availability of new crop technology’, ‘import of Maize’ and ‘stopped intercropping’ shows negative association with farm dynamics. These variables are hindering the production of Maize and areal increase of its cultivation. Whereas ‘Change in fertilizer use’, ‘cultivated area increase’ and ‘started selling maize’ are the factors supporting the argument of areal cultivation and Maize production increase. The most important explanation of farm dynamism is described by farm holders’ access to modern crop technologies, and commercial incentives by the government to staple crop production.