Spatial dependence of production choice: application of Bayesian Spatial Autoregressive Probit model on smallholder rubber producers

Jagath Chaminda Edirisinghe, Keminda Herath, Udith Jayasinghe-Mudalige, Sachintha Mendis


This research is conducted to study the output choice of rubber farmers with the special interest in measuring the neighbourhood effect in the choice production choice. We make inferences about the choice of production of Ribbed Smoked Sheets (RSS) for selling. In doing so, we estimate a Spatial Autoregressive Probit Model (SAPM) using recently developed Bayesian techniques in estimating the SAPM. We find that significant spatial relations exist in the production of sheet rubber and there are considerable amounts of spillover effects. The overarching issues of product choice are the availability of physical resources and the lack of proper grading system. Our findings reveals the possibility of central processing to overcome resource limitations, significant reductions in extension efforts in promoting good manufacturing practices by taking stock of the ‘neighbourhood’ effect present in farmer choices and the necessity of proper grading and quality management system to win producer trust.

Full Text: