Routine to obtain several goodness-of-fit statistics for spatial and non-spatial models

Source: R/asreml_funct.R

stats.spatial Internal routine that calculates goodness-of-fit statisitcs for a specific fitted model provided as input. Statistics obtained are log-likelihood, AIC, BIC, heritability based on predictor error variance (heritPEV, only for genotypes 'random'), A-optimality value, and logarithm of D-optimality value,

stats.spatial(
  object = NULL,
  checks = FALSE,
  solution = NULL,
  preds = NULL,
  k = 1,
  ctable = NULL
)

Arguments

object

model fit output for object class asreml. Genotype factor is called 'gen'

checks

logical to indicate if checks were included in the model (default = FALSE)

solution

data frame with the random solutions from asreml (BLUPs)

preds

data frame with the predictions for genotypes (required to calculate Aopt, Dopt)

k

value 1 or 2 to be used for the denominator of heritabilty calculation (herit.PEV) (default = 2)

ctable

table with genotypes and checks (optional)

Value

A table with goodness-of-fit statistics for models evaluated. This includes columns: number of variance components in the model (n.VC), log-likelihood (logL), Akaike information criteria (AIC), Bayesian information criteria (BIC), heritability based on variance components, heritability based on predictor error variance (heritPEV, only for genotypes 'random'), A-optimality value, and logarithm of D-optimality value.

Author

Salvador A. Gezan. VSN International

Examples

# Example: Pending