Bayesian Analysis of Spatial Probit Models in Wheat Waste Management Adoption

The purpose of this study was to identify factors influencing the adoption of wheat waste management by wheat farmers. The method used in this study using the spatial Probit models and Bayesian model was used to estimate the model. MATLAB software was used in this study. The data of 220 wheat farmers in Khouzestan Province based on random sampling were collected in winter 2016. To calculate Bayesian coefficients the Gibbs sampling and Metropolis–Hastings algorithm were used. A Lagrange Multiplier test for spatial error dependence [LM(err)] and a Lagrange Multiplier test for spatial lag dependence [LM(lag)] to extract the appropriate model were used.The results of both models were statistically significant with 99% probability. Thus, both models can be used in interpreting the results. Based on the results of the estimation of spatial models the variables of participation in extension courses, technical knowledge about management of waste, income, crop’s yield, mechanization level and the spatial autoregressive coefficient had significant role on adoption of waste management.