Validation of Growth Rate Estimate Using Spline Regression Technique for Green Gram Production in Odisha

Abstract

The varying pattern of agriculture data in different phases over a long period could be thought of capturing effectively by the spline regression technique. The present study focuses on using spline regression in estimating the growth rate of production of green gram for Odisha. This technique requires the segmentation of the entire study period into distinct periods by inserting knots at suitable intervals. It ensures the same pattern of variation in a particular segment and an abrupt change in the pattern of variation in the next segment. The spline regression approach was compared to the conventional regression technique based on specific statistical criteria to confirm its performance. To estimate the growth rate of green gram in Odisha, spline regression models outperformed conventional regression models, which is why they were chosen for the first and last segments of the study period for area, yield, and output. Keywords: Continuity, growth rate, knot, segment, spline regression. JEL Codes: B23, C32, G21.