Abstract

PEST is a general purpose Fortran package for multivariate prediction and estimation. It has no inherent limits on number of traits, effects (fixed or random), and covariables. Interactions and polynomials are supported. It further features multiple incidence matrices in multivariate models, heterogeneous residual variances, missing values, inclusion of relationship information with and without inbreeding and genetic group models, hypothesis testing, loading of old solutions, and reading of data from memory or disk. PEST supports 5 solvers which may be combined to perform a two level block iteration. Equation systems are either set up in memory or solved by iteration on data or a combination of both allowing to choose the most efficient procedure relative to the size of the problem. As PEST operates on original codes of up to 16 characters length, no data preparation is required. PEST is driven by a parameter file.
 

Eildert Groeneveld, M. Kovac, T. Wang

Proceedings of the World Congress on Genetics Applied to Livestock Production, Volume XIII. Plenary lectures, molecular genetics and mapping, selection, prediction and estimation., , 488–491, 1990
Download Full PDF BibTEX Citation Endnote Citation Search the Proceedings



Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.