June 2007
Beginner to intermediate
950 pages
27h 8m
English
The following experiment, with Weight as the response variable, involved Genotype and Sex as two categorical explanatory variables and Age as a continuous covariate. There are six levels of Genotype and two levels of Sex.
Gain <-read.table("c:\\temp\\Gain.txt",header=T) attach(Gain) names(Gain) [1] "Weight" "Sex" "Age" "Genotype" "Score"
We begin by fitting the maximal model with its 24 parameters: different slopes and intercepts for every combination of Sex and Genotype.
m1<-lm(Weight~Sex*Age*Genotype) summary(m1)
Coefficients: Estimate Std. Error t value Pr (>|t|) (Intercept) 7.80053 0.24941 31.276 < 2e-16 *** Sexmale -0.51966 0.35272 -1.473 0.14936 Age 0.34950 0.07520 4.648 4.39e-05 *** GenotypeCloneB 1.19870 0.35272 3.398 0.00167 ** GenotypeCloneC -0.41751 0.35272 -1.184 0.24429 GenotypeCloneD 0.95600 0.35272 2.710 0.01023 * GenotypeCloneE -0.81604 0.35272 -2.314 0.02651 * GenotypeCloneF 1.66851 0.35272 4.730 3.41e-05 *** Sexmale:Age -0.11283 0.10635 -1.061 0.29579 Sexmale:GenotypeCloneB -0.31716 0.49882 -0.636 0.52891 Sexmale:GenotypeCloneC -1.06234 0.49882 -2.130 0.04010 * Sexmale:GenotypeCloneD -0.73547 0.49882 -1.474 0.14906 Sexmale:GenotypeCloneE -0.28533 0.49882 -0.572 0.57087 Sexmale:GenotypeCloneF -0.19839 0.49882 -0.398 0.69319 Age:GenotypeCloneB -0.10146 0.10635 -0.954 0.34643 Age:GenotypeCloneC -0.20825 0.10635 -1.958 0.05799 . Age:GenotypeCloneD -0.01757 0.10635 -0.165 0.86970 Age:GenotypeCloneE ...
Read now
Unlock full access