Analysis of test day milk yield by random regression models and evaluation of persistency in Iranian dairy cows.

Variace / covariance components of 227118 first lactaiom test-day milk yield records belonged to 31258 Iranian Holstein cows were estimated using nine random regression models. Afterwards, different measures of persistency based on estimation breeding value were evaluated. Three functions were used to adjust fixed lactation curve: Ali and Schaeffer (AS), quadratic (LE3) and cubic (LE4) order of Legendre polynomial but for random effects, unequal order of Legendre polynomials (LE3, LE4, LE5 and Ali and Schaffer) functions were evaluated. Heterogeneous residual variance considered during days in milk and evaluation of models was based on eigenvalues and associated eigenvectors and residual variance. Model with Ali and Schaeffer function for fixed part and LE3 and LE4 for additive and permanent environmental effects was selected as the best model for random regression analysis in first parity dairy cows. The highest and lowest heritability were observed in the middle (0.29) and beginning (0.08) of lactation, respectively. Persistency measurement proposed by Cobuci (PSY1) (difference between estimation breeding value between 290 and 90 days) was preferential for using in further genetic evaluations for persistency in milk yield of Iranian Holstein cows.