Nonlinear growth models for describing country’s Lentil (Lens culinaris M.) production

Abstract

Nonlinear statistical growth models play a very important role in understanding the dynamics of an agricultural system. Different nonlinea r me cha nistic g rowth mo dels vi z, monomolecular, logistic and Gompertz models were used to describe the country’s lentil (Lens culinaris M.) production during 1970-71 to 2009-10. The parameters of these models were estimated using Levenberg-Marquardt algorithm for nonlinear estimation. Gompertz model was found to be most suitable for this dataset based on various goodness of fit criteria viz.Coefficient of determination ( )2 R1, Root mean squared error, Mean absolute error, Correlation between observed and predicted valuesand Percentage forecast error. The predicted lentil production for the year 2009-10 was 1.02 million tonneswhereas the actual production was 1.03 million tonnes. From this study it was concluded that lentil production in India reached about 87% of its potential (with present technological status). So with around 5% intrinsic growth rate it has ample scope in increasing production in near future.