Logistic regression analysis for prediction of stayability of Karan Fries crossbred cattle

Abstract

A dairy cow with poor fertility is likely to be culled voluntarily from the herd thus affecting the overall profitability of the farm. Standardization of voluntary waiting period (VWP) would help in making important managerial decisions regarding post-partum breeding of the animals and estimating pregnancy rate. Present investigation was carried out with the purpose of VWP standardization for daughter pregnancy rate (DPR) estimation and prediction of stayability upto 3 lactations (STAY3) and 4 lactations (STAY4) on the basis of first lactation 305 day milk yield (F305MY) and first lactation daughter pregnancy rate (FDPR) performance using logistic regression (LR). ROC (receiver operating characteristic) analysis revealed that the area under curve of 65.44% for STAY3 and 58.48% for STAY4. Odd ratio estimates indicated that the cows with production performance around 3661.10 kg had 2.67 and 1.60 times greater chance of stayability than cows whose production level was around 2677.50 kg and 3133.70 kg, respectively. The odds ratio estimates indicated that the cows with higher production performance had greater chance of staying in the herd in comparison to cows with superior fertility performance. LR analysis helped in identifying the optimum levels of first lactation fertility and production performance levels for Karan Fries cattle.