Prediction of Lifetime Milk Yield by Using Penalized Regression in Crossbred Cattle

Penalized regression have been proposedto improve OLS estimates. Ridge regression and LASSO (least absolute shrinkage and selection operator) minimize the residual sum of squares subject to a penalty of the L2-norm and L2-norm respectively on the regression coefficients. Elastic net, has combination of L1 and L2 penalties. We predict the milk production and assessed the relative performance of the above three penalized model. Based on R², RMSE, MAE, MAPE and U statistics it is found that elastic net technique performed better.