Score.BinReg: Score Test for Binary Regression
In zrmacc/BinReg: BinReg

Description Usage Arguments Value Examples

Tests the hypothesis that a subset of the regression coefficients are fixed at a reference value. Specifically, let β denote the regression coefficient. Partition β=(β_{1},β_{2}). Suppose that interest lies in testing that β_{1} is fixed at β_{10}. Score.BinReg performs a score test of H_{0}:β_{1}=β_{10}. The test is specified using a logical vector L, with as many entries as columns in the model matrix X. The values of L set to T are constrained under the null, while values of L set to F are estimated under the null.

1 2	Score.BinReg(y, X, L, b10 = NULL, model = "logistic", df = NULL, sig = 0.05, eps = 1e-08, maxit = 10, report = T)

`y`	Numeric response vector.
`X`	Numeric model matrix.
`L`	Logical vector, with as many entires as columns in the model matrix, indicating which columns have fixed coefficients under the null.
`b10`	Value of the regression coefficient for the selected columns under the null. Defaults to zero.
`model`	Selected from among logistic, probit, and robit.
`df`	Degrees of freedom, if using the robit model.
`sig`	Significance level, for CIs.
`eps`	Tolerance for Newton-Raphson iterations.
`maxit`	Maximum number of NR iterations.
`report`	Report fitting progress?

A numeric vector containing the score statistic, the degrees of freedom, and a p-value.

## Not run: 
set.seed(101);
# Design matrix
X = cbind(1,matrix(rnorm(n=4*1e3),nrow=1e3));
# Regression coefficient
b = c(1,-1,2,-1,0);
# Logistic outcome
y = rBinReg(X,b,model="logistic");
# Test b1=b2=b3=b4=0, which is false.
Score.BinReg(y=y,X=X,L=c(F,T,T,T,T),model="logistic",report=F);
# Test b4=0, which is true.
Score.BinReg(y=y,X=X,L=c(F,F,F,F,T),model="logistic",report=F);
# Test b2=0 and b4=2, which if false.
Score.BinReg(y=y,X=X,L=c(F,F,T,F,T),b10=c(0,2),model="logistic",report=F);
# Test b1=b3=-1, which is true. 
Score.BinReg(y=y,X=X,L=c(F,T,F,T,F),b10=c(-1,-1),model="logistic",report=F);

## End(Not run)