ICSKAT (version 0.3.0)

ICSKATwrapper: ICSKATwrapper.R

Description

Wrapper to fit the null model and run ICSKAT all in one instead of separately - offers some functionality for error checking or using different initial values when fit fails to converge.

Usage

ICSKATwrapper(
  left_dmat,
  right_dmat,
  initValues,
  lt,
  rt,
  obs_ind,
  tpos_ind,
  gMat,
  PH = TRUE,
  nKnots = 1,
  maxIter = 3,
  eps = 10^(-6),
  runOnce = FALSE,
  returnNull = FALSE
)

Value

Either a list with skatOutput and nullFit (two lists), or just skatOutput.

Arguments

left_dmat

n*(p+nknots+2) design matrix for left end of interval.

right_dmat

n*(p+nknots+2) design matrix for right end of interval.

initValues

(p+nknots+2)*1 vector of coefficients to initialize the Newton-Raphson.

lt

Left side of interval times.

rt

Right side of interval times.

obs_ind

n*1 vector of whether the event was observed before last follow-up.

tpos_ind

n*1 vector of whether the event was observed after follow-up started (t>0).

gMat

n*q matrix of genotypes.

PH

Boolean for whether to fit PH model (TRUE) or PO model (FALSE).

nKnots

Number of knots in the spline.

maxIter

Number of times to try the fit if initial values do not lead to convergence.

eps

Difference in L2 norm of fitted null coefficients that stops the Newton Raphson.

runOnce

Boolean, if true then just go through the algorithm once with the initial values for coefficients, updating the variance matrix, useful for bootstrapping.

returnNull

Return a list with the skat output and null model, or just return the skat output (FALSE).

Examples

set.seed(0)
gMat <- matrix(data=rbinom(n=2000, size=2, prob=0.3), nrow=100)
xMat <- matrix(data=rnorm(200), nrow=100)
bhFunInv <- function(x) {x}
obsTimes <- 1:5
etaVec <- rep(0, 100)
outcomeDat <- gen_IC_data(bhFunInv = bhFunInv, obsTimes = obsTimes, windowHalf = 0.1,
probMiss = 0.1, etaVec = etaVec)
lt <- outcomeDat$leftTimes
rt <- outcomeDat$rightTimes
tpos_ind <- as.numeric(lt > 0)
obs_ind <- as.numeric(rt != Inf)
dmats <- make_IC_dmat(xMat, lt, rt, obs_ind, tpos_ind)
ICSKATwrapper(left_dmat = dmats$left_dmat, right_dmat = dmats$right_dmat,
initValues = rep(0, ncol(xMat) + 3), lt = lt, rt = rt, obs_ind = obs_ind,
tpos_ind = tpos_ind, gMat = gMat, returnNull = TRUE)