Estimation and prediction problems in mixed linear models for maternal genetic effects
Cantet, Rodolfo Juan Carlos
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/23630
Description
Title
Estimation and prediction problems in mixed linear models for maternal genetic effects
Author(s)
Cantet, Rodolfo Juan Carlos
Issue Date
1990
Doctoral Committee Chair(s)
Gianola, Daniel
Department of Study
Biology, Genetics
Statistics
Agriculture, Animal Culture and Nutrition
Discipline
Biology, Genetics
Statistics
Agriculture, Animal Culture and Nutrition
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Biology, Genetics
Statistics
Agriculture, Animal Culture and Nutrition
Language
eng
Abstract
The main objectives of this study were: (1) to compare quadratic and likelihood estimators of dispersion parameters in a sire plus maternal grandsire model (S-M) with respect to estimated bias (EB) and estimated mean squared error (EMSE) when the data are affected by selection, (2) to estimate direct and maternal (co)variance components (CVC) using an animal model (MAM) and genetic grouping for weaning weight in beef cattle, (3) to extend the theory of genetic grouping in models with maternal effects allowing for differential criteria to be used when assigning groups for direct and maternal effects, and (4) to present a Bayesian approach to estimation of CVC in MAM. To achieve objective (1) some designs based on S-M were compared with respect to the variance of estimates of heritability. Also, simulated data under S-M and either: (1) random, (2) translation invariant, or (3) location-variant selection were employed. Although perhaps the model and design did not permit to reveal large differences in EB and EMSE among estimators, likelihood based methods tended to outperform quadratic ones with respect to EMSE and, to a lesser extent, with respect to EB. This was specially so under non-random selection. Records from 935 Angus calves were used to estimate CVC in MAM by restricted maximum likelihood to attain objective (2). Estimates of CVC and of functions thereof did not differ very much in models which included or excluded genetic groups. Estimates of heritability for direct and maternal effects were smaller than those reported previously. The estimate of the additive correlation between direct and maternal effects was $-$0.31. Genetic and environmental trends for direct effects were positive whereas corresponding trends for maternal effects were close to zero. The Bayesian approach for making inferences about CVC in MAM used inverted-Wishart and inverted chi-square prior distributions. The joint posterior density of CVC was obtained in closed form and three methods to achieve further marginalization were discussed.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
