A simulation study was carried out to investigate the effect of selection o
n the estimation of genetic parameters for butterfat production in dairy ca
ttle. It was found that selection leads to a substantial overestimation of
fixed effects and variance components and an underestimation of heritabilit
y estimates. This effect was attributed to the fact that all information on
all animals in the analysis should be available to compensate for the effe
ct of selection.