tcl_scorefun: Computation of score funtion.

In akurz1/tcl: Testing in Conditional Likelihood Context

Description

Uses function jacobian() from numDeriv package to compute (approximate numerically) score function (first order partial derivatives of conditional log likelihood function) evaluated at arbitrary values of item easiness parameters.

Usage

tcl_scorefun(X, eta, W, model = "RM")

Arguments

X	data matrix.
eta	numeric vector of item easiness parameters.
W	design matrix.
model	RM, PCM, RSM, LLTM.

Value

Score function evaluated at eta

References

Gilbert, P., Gilbert, M. P., & Varadhan, R. (2016). numDeriv: Accurate Numerical Derivatives. R package version 2016.8-1.1. url: https://CRAN.R-project.org/package=numDeriv

Examples

# Rasch model with beta_1 restricted to 0
y <- eRm::raschdat1
res <- eRm::RM(X = y, sum0 = FALSE)
scorefun <- tcl_scorefun(X = y, eta = res$etapar, model = "RM")

akurz1/tcl documentation built on Oct. 22, 2021, 4:10 p.m.