hdslcse.fast {hds}

Hazard discrimination summary (local constant) standard error estimate

Description

hdslcse.fast calculates an estimate of the variance for the local constant hazard discrimination summary estimator at a time t. The time t is implied by S, betahat, and betahatse

Usage

hdslcse.fast(S, betahat, m, betahatse)

Arguments

S

A vector of length nrow(m) (which is typically the number of observations n), where each value is the subject-specific survival at time t where t is implied by the choice of betahat.

betahat

A p x 1 vector of coefficient estimates at time t of interest from the local-in-time Cox model. Vector length p is the number of covariates. Typically the output from hdslc::finda is passed here.

m

A numeric n x p matrix of covariate values, with a column for each covariate and each observation is on a separate row.

betahatse

A p x p covariance matrix for betahat at time t

Details

The use will typically not interact with this function directly. Instead this function is wrapped by hdslc.

Value

Variance estimate that has not been normalized. To get a usable standard error estimate, divide the output of this function by the bandwidth and sample size, and then take the square root.