Factors affecting farmer's choice of cultivating landrace rice: Using a switching regression model

Abstract

© 2016 by the Mathematical Association of Thailand. All rights reserved. The study aim to investigate the inuential factors of the farmer's decision to cultivate landrace rice. To achieve our study, a switching regression model which contain two models, selection and outcome models is used as a tool. The Bayesian method is employed to joint the distribution of error terms of the selection and outcome models under the six copula functions. Results show that the selection model has the normal distribution while the outcome model has the logistic distribution. The Clayton copula is best fit to the farmer’s decision model. Overall, the more educated farmer is 4.7% less likely to grow landrace rice. The income from non-agricultural activities has slightly effect on the farmer’s decision. Further, the household with available water supply is 5.4% more likely to reduce the cultivated area of the landrace rice in the next season. This because of the local rice varieties is drought tolerant; therefore, the farmers would prefer to use available water for planting other crops that give a higher return.