Assessing relationships between genetic evaluations using robust regression with an application to Holsteins in Uruguay

Globalization of dairy cattle breeding has created a need for international sire proofs. Some early methods for converting proofs from one population to another are based on simple linear regression. An alternative robust reg ression method based on the t-distribution is presented, and maximum likeli hood and Bayesian techniques for analysis are described, including the situ ation in which some proofs are missing. Procedures were used to investigate the relationship between Holstein sire proofs obtained by two Uruguayan ge netic evaluation programs. The results suggest that conversion equations de veloped from data including only sires having proofs in both populations ca n lead to distorted results, relative to estimates obtained using technique s for incomplete data. There was evidence of non-normality of regression re siduals, which constitutes an additional source of bias. A robust estimator may not solve all problems, but can provide simple conversion equations th at are less sensitive to outlying proofs and to departures from assumptions .