varEst: Estimate the Frailty Variance for Discrete-Time Multivariate...

In lclGWAS: Efficient Estimation of Discrete-Time Multivariate Frailty Model Using Exact Likelihood Function for Grouped Survival Data

Description

A method to estimate the frailty variance for a multivariate frailty model accounting for family structure of related individuals (i.e., trios). The input data is assumed to be organized such that records for each family occur consecutively, and that records for offspring precede those for parents. The variance matrix for the random effects is assumed to be of the form var*K, where K is a matrix of kinship coefficients between family members. The following groupings are permitted: (Individual), (Offspring, Offspring), (Offspring, Parent), (Offspring, Parent, Parent), and (Offspring, Offspring, Parent, Parent). Other family structures have not been implemented.

Usage

  varEst(fam_group, alpha, dtime, delta, g, beta, lower, upper)  	

Arguments

fam_group	Vector of family indicators for each sample.
alpha	Vector of baseline survival rates for each time interval.
dtime	Vector of observed discrete survival times for each sample.
delta	Event indicator vector: 1 indicates observed event, 0 indicates censored.
g	Vector of numeric genotypes for each sample.
beta	Scalar for the fixed effect size.
lower	Scalar for the lower bound of the variance estimation search region.
upper	Scalar for the upper bound of the variance estimation search region.

Value

A list with one object:

varEst

Scalar estimate of the frailty variance.

Examples

# Generate dummy data	
fam_group <- rep(1:3,each=3)
alpha <- c(0.7500000, 0.6666667, 0.5000000, 0.0000000)
dtime <- c(1, 3, 3, 2, 1, 1, 2, 3, 1)
delta <- c(1, 0, 1, 1, 1, 0, 1, 0, 1)
g     <- c(0, 1, 1, 1, 2, 2, 0, 0, 0)
beta  <- 1
res   <- varEst(fam_group, alpha, dtime, delta, g, beta, lower = 0, upper = 2) 
res

lclGWAS documentation built on May 2, 2019, 2:46 p.m.