Listy Biometryczne - Biometrical Letters Vol. 32(1995), No. 2, 51-60


Show full-size cover
ESTIMATION OF PARAMETERS IN LINEAR MODEL
WITH THE PRESENCE OF SELECTION

J. Hyanek1, K. Moliński2 and T. Szwaczkowski3

1 Research Institute of Animal Production, 10400 Praha-Uhri neves,
Czech Republic
2 Department of Mathematical and Statistical Methods, Agricultural
University, Wojska Polskiego 28, 60-637 Poznan, Poland
3Department of Genetics and Animal Breeding, Agricultural University,
Wolynska 33, 60-637 Poznan, Poland


The paper contains the theoretical considerations on an estimation of parameters in linear model when original data are selected by given truncated point, but the selection intensity is not known. The maximum likelihood estimators in truncated normal distribution are determined. In the next part of the study, the estimation of parameters in mixed model (under truncated normal distribution) is discussed.