effect_null.manyglm {ecopower}

Specify null effects for multivariate abundance data

Description

effect_null returns a coefficient matrix to be parsed to powersim by default to specify a null effect.

Usage

## S3 method for class 'manyglm'
effect_null(object, term)

effect_null(object, term)

Arguments

object

objects of class manyglm, typically the result of a call to manyglm.

term

Name of predictor of interest in quotes.

Details

effect_null produces a coefficient matrix with a null effect that is specified by setting the parameter estimates of a predictor of interest term to 0. This function is used by default in powersim. Note that intercept values are parameterised as in effect_alt.

Value

A coefficient matrix with the null effect.

Functions

• effect_null(): Specify null effects for multivariate abundance data

See Also

effect_alt, powersim

Examples

library(mvabund)
data(spider)
spiddat = mvabund(spider$abund)
X = data.frame(spider$x)

# Find null effect size for continuous predictor
spid.glm = manyglm(spiddat~soil.dry, family="negative.binomial", data=X)
coeffs0 = effect_null(spid.glm, term="soil.dry")