Identifying environmental variables explaining genotype-by-environment interaction for body weight of rainbow trout (Onchorynchus mykiss): reaction norm and factor analytic models

Table 5 Model comparison between five different reaction norm models, factor analytic model, and multivariate model on body weight at harvest

Model	EV	LogL	NPar	AIC	BIC
Reaction norm	Age	0.0	7	87721.5	87735.1
	Temperature	3.1	7	87715.3	87728.9
	Day*Degree	32.5	7	87656.5	87670.2
	Oxygen	38.8	7	87643.8	87657.5
	Photoperiod	43.0	7	87635.4	87649.0
Factor analytic	Latent	105.2	12	87521.0	87544.4
Multivariate	N.A.	114.12	14	87507.2	87534.5

EV = environmental variable; LogL = natural logarithm of likelihood deviated from the smallest value (Age: -43853.7); NPar = number of parameters; AIC = Akaike’s information criterion, BIC = Bayesian’s information criterion; bold letter indicates the lowest AIC and BIC from both random regression and factor analytic models; residual degrees of freedom are equal to 8854 (reaction norm and factor analytic model) and 8853 (multivariate model); N.A. = not applicable.

ISSN: 1297-9686