A comparison of three methods for variance components estimation to compute the individual weight in a kilogram of male and female albino rats placed in five different cages after feeding them for a week in the Animal farmhouse of Olabisi Onabanjo University, Ago-Iwoye, Ogun State, Nigeria. An analog weighing scale was used to measure the weight, and measurements were taken five times during the week-long experiment. Analysis of variance (ANOVA), Maximum likelihood (ML), and Restricted maximum likelihood (REML) estimations were applied to assess the accuracy and reliability of the weight measurements. However, the comparison shows that estimates of ML and REML are the same, the estimates of the three methods used are not within the same parameter space. It was concluded that for a balanced one-way model, the ANOVA method is better for estimating variance in animals for both Male and Female.
Key words: Analysis of variance (ANOVA), Maximum Likelihood (ML), Restricted Maximum Likelihood Estimation (RMLE), Variance components, Parameter space
|
