backward: Inference based model selecton for IRT models.

In nando11235813/raschreg: Item Response Theory models including explanatory variables of the ability parameter

Description

Select the optimal number of variables by means of LRT or F tests.

Usage

backward(mod, level = 0.05, test = 'LRT', cov_type = 'hessian')

Arguments

mod	A fitted rasch model.
level	Significance level to bu used.
test	A character string, matchiing one of LRT o Wald.
cov_type	A character indicating which covariance matrix estimation is to be used. Possible choices are 'hessian', 'opg' and 'sandwich'.

Details

backward uses drop1 repeatedly until all explanatory variables in the model are significant at a specified level.

Value

backward returns an object of class rasch containing the optimal subset of variables.

Author(s)

Fernando Massa, [e-mail]

See Also

drop1

nando11235813/raschreg documentation built on Oct. 2, 2021, 3:11 p.m.