DEVELOPMENT OF A MATHEMATICAL MODEL FOR UPLAND RICE PRODUCTION (A CASE STUDY OF NATIONAL CEREALS RESEARCH INSTITUTE BADDEGI, NIGERIA).

Abstract

In this paper, mathematical development of two quadratic models was used to obtain minimum, maximum and saddle points of yield response of upland rice production. The results showed that the models were adequate and significant at 5% by using canonical analysis through the method of least square, R2 (coefficient of determination) strength, R2 –adjust, coefficient of variation (CV) and root mean square error (RMSE). The results also indicated that the quadratic effect of irrigation is important during the dry season than nitrogen for optimum yields. It was also established that there was an increase in yield when variance (ANOVA) was used for the data collected from National Cereal Research Institute Baddegi, Niger State. And computation of the data was adequate with R2 above 60%, R2 –adjusted above 55% and RMSE was very small.