raschsk: Fitting Asymmetric Rasch model

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

Description

Estimate the asymmetric Rasch model approximating the marginal log-likelihood using Laplace approximation.

Usage

raschsk(items, init = NULL)

Arguments

items	Numeric matrix (or data.frame) containing only zeroes and ones (J columns).
init	Optional numeric vector of initial values at which initialize numerical optimization (length J + 1).

Details

The asymmetric model differs from the original in that it assumes that ability is distributed Skew-Normal (following the centred parameters form described by Azzalini). Optimizes raschlikLA numericaly via nlminb. Rows containing at least one NA are removed from items. Standard errors of model parameters are approximated by inverting the observed information matrix.

Value

An object of class rasch is a list containing the following componentes:

call	The matched call
coef	A named vector of coefficients
iter	Number of iterations used to optimize de log-likelihood
loglik	The log-likelihood value
vcov	The variance-covariance matrix of the estimated parameters
items	The item matrix
beta	(Only when regression terms are included) the estimated regression parameters
linpred	(Only when regression terms are included) prediction covariates

Author(s)

Fernando Massa, [e-mail]

References

\insertRef

rasch1960raschreg

\insertRef

baker2004raschreg

\insertRef

azzalini2013raschreg

See Also

rasch, raschd, raschreg, raschdreg, irt2p, irt2preg

Examples

n   <- 100
X   <- sim_rasch(n)
mod <- raschsk(X)

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