Genetic evaluation for production traits in the Holstein breed in Italy is based on a Random Regression Test Day Model (RRTDM) since November 2004. More specifically the model is a multiple lactation, multiple traits RRTDM, similar to the model used in Canada for official genetic evaluation. In order to improve stability, data were pre-adjusted for number of days pregnancy at test day and fixed regression curve effects include now the effect of year of production, four calving seasons instead of two and a general effect for days in milk (DIM) was added. The changes in the model improved stability of proofs over time, thus increasing the ability of the test day to accurately predict the breeding value of both bulls and cows. Trend validation results showed also improvements and residuals were positively affected by the changes. Trends in trait residuals previously detected with respect to DIM and month of calving did show a significant reduction. The new model was officially introduced in January 2009.
Improving the stability of test day model evaluation for production traits in the Italian Holstein / Canavesi, F; Biffani, S; Bramante, G; Finocchiaro, R. - In: ITALIAN JOURNAL OF ANIMAL SCIENCE. - ISSN 1594-4077. - 8:(2009), pp. 39-41. [10.4081/ijas.2009.s2.39]
Improving the stability of test day model evaluation for production traits in the Italian Holstein
Biffani S;
2009-01-01
Abstract
Genetic evaluation for production traits in the Holstein breed in Italy is based on a Random Regression Test Day Model (RRTDM) since November 2004. More specifically the model is a multiple lactation, multiple traits RRTDM, similar to the model used in Canada for official genetic evaluation. In order to improve stability, data were pre-adjusted for number of days pregnancy at test day and fixed regression curve effects include now the effect of year of production, four calving seasons instead of two and a general effect for days in milk (DIM) was added. The changes in the model improved stability of proofs over time, thus increasing the ability of the test day to accurately predict the breeding value of both bulls and cows. Trend validation results showed also improvements and residuals were positively affected by the changes. Trends in trait residuals previously detected with respect to DIM and month of calving did show a significant reduction. The new model was officially introduced in January 2009.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.