Trend analysis of India’s agricultural exports using linear and nonlinear growth models

Abstract

This study analyzes the long-term trend of India’s agricultural exports using twelve linear and nonlinear growth models applied to annual export data. The models tested include linear, logarithmic, inverse, quadratic, cubic, compound, power, S-curve, growth, exponential, logistic, and Gompertz forms. Model performance was compared using Adjusted R², RMSE, MAPE, AIC, BIC, and Theil’s U. The results show that export behavior is nonlinear, and the logistic model gives the best fit. It recorded the highest Adjusted R² (0.9176) and the lowest RMSE (38,834.56), MAPE (26.68), and Theil’s U (0.0999). The logistic curve shows an S-shaped trend in which exports grow slowly at first, rise faster later, and then move toward a saturation point. Applying the same method to India’s total exports also identified the logistic model as the best. Four-year forecasts show steady growth in agricultural exports and faster growth in total exports, supporting the usefulness of nonlinear models for policy decisions.