Comparison of Lasso and stepwise regression technique for wheat yield prediction

Multiple regression approach has been used to forecast the crop production widely. This study undertaken evaluate performance of stepwise and Lasso (Least absolute shrinkage selection operator) technique in variable development wheat model for yield using weather data period 1984-2015, collected from IARI, New Delhi. Statistical parameters viz. R2, RMSE, MAPE were 0.81, 195.90 4.54 per cent respectively with 0.95, 99.27, 2.7 percentage, regression. Forecast models validated during 2013-14 2014-15. Prediction errors -8.5 10.14 1.89 1.64 percent Lasso. shows that is better than some extent.