logl.test: Log-likelihood test for comparing terms in ASReml-R models

In biometryhub/BiometryTraining: Functions to Assist Design and Analysis of Agronomic Experiments

Description

Log-likelihood test for comparing terms in ASReml-R models

Usage

logl.test(
  model.obj,
  rand.terms = NULL,
  resid.terms = NULL,
  decimals = 3,
  numeric = FALSE,
  quiet = FALSE
)

Arguments

model.obj	An ASReml-R model object
rand.terms	Random terms from the model. Default is NULL.
resid.terms	Residual terms from the model. Default is NULL.
decimals	Controls rounding of decimal places in output. Default is 3 decimal places.
numeric	Return p-values as numeric? Default is that they are characters, where very small values shown as less than a small number. See details for more.
quiet	Logical (default: FALSE). Hide warnings and messages?

Details

Typically p-values cannot be 0, and are usually just below some threshold of accuracy in calculation of probability.

Value

A dataframe containing the results of the test.

Examples

## Not run: 
library(asreml)
dat <- asreml::oats
dat <- dat[order(dat$Row, dat$Column),]

#Fit ASReml Model
model.asr <- asreml(yield ~ Nitrogen + Variety + Nitrogen:Variety,
                    random = ~ Blocks + Blocks:Wplots,
                    residual = ~ ar1(Row):ar1(Column),
                    data = dat)
oats.logl <- logl.test(
  model.obj = model.asr, rand.terms = c("Blocks", "Blocks:Wplots"),
  resid.terms = c("ar1(Row)", "ar1(Column)")
)
oats.logl

## End(Not run)

biometryhub/BiometryTraining documentation built on Jan. 27, 2022, 7:52 p.m.