MULTICOLLINEARITY DIAGNOSTICS UPON COBB-DOUGLAS PRODUCTION FUNCTION FOR ESTIMATING RESOURCE USE EFFICIENCY OF TOMATO IN CHITWAN, NEPAL

Abstract

Testing the multicollinearity problem upon the empirical estimation of the Cobb-Douglas production function forms a major part of the research analysis, but many analysts avoid it, which may cause a statistical error on the regression coefficients and correlation. Keeping this point in view, a study was carried out in Chitwan district in 2020 to analyze the profitability and resource use efficiency of tomato production with testing of collinearity problem on regression analysis. A total of 180 tomato growers were selected based on Cochran’s formula of sample size determination, for interviews that used a pretested questionnaire. The benefit-cost ratio in the research area was 1.28 which indicates that tomato production is profitable and farmers of Chitwan get an additional 28 paisa with an investment of one rupee in tomato farming. The cost for tillage, potash, seed, and labor is overutilized, while for FYM, DAP, urea, and irrigation it is underutilized. The empirical findings of collinearity diagnostics revealed that there is no multicollinearity problem in the present regression analysis. Thus, the Cobb-Douglas production function was used without the need for any correction. The analysis suggests that the efficient utilization of resources is necessary to increase the profitability of tomato-growing farmers in the study area.

Keywords

• profitability
• variance inflation factors
• tolerance
• b/c ratio
• regression