CDsampling (version 0.1.6)

Fi_func_MLM: Generate Fisher information matrix F_x at a design point x_i for Multinomial logistic regression model

Description

Generate Fisher information matrix F_x at a design point x_i for Multinomial logistic regression model

Usage

Fi_func_MLM(X_x, beta, link)

Value

F_x is the Fisher information matrix at design point x_i (with model matrix X_x);

U_x is a middle step matrix for calculation of F_x, details see Corollary 3.1 in Bu, X., Majumdar, D., & Yang, J. (2020). D-optimal designs for multinomial logistic models

Arguments

X_x

model matrix given design point x_i (for example, X_x = h.func(x_i), where h.func transforms a design point to a model matrix)

beta

parameter coefficients in the Multinomial logistic regression model, the order of coefficients in bvec and the order of design points in X_x should be consistent

link

link function of Multinomial logistic regression model, options are "baseline", "cumulative", "adjacent", or "continuation"

Examples

X_x_temp = rbind(c( 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0),
                c( 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0),
                c( 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0),
                c( 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0),
                c( 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0))
bvec_temp = c(-4.047, -2.225, -0.302, 1.386, 4.214, 3.519,
2.420, 1.284, -0.131, -0.376, -0.237, -0.120)
link_temp = "cumulative"
Fi_func_MLM(X_x=X_x_temp, beta=bvec_temp, link=link_temp)