CoxBcv.mdmr {CoxBcv} | R Documentation |
Hybrid MDMR bias-corrected sandwich variance estimator
Description
Calculate the hybrid MDMR bias-corrected sandwich variance estimator, for marginal Cox analysis of cluster randomized trials, proposed by Wang et al. (under review). MD: Mancl and DeRouen (2001); MR: martingale residual.
Usage
CoxBcv.mdmr(Y, Delta, X, ID)
Arguments
Y |
vector of observed time-to-event data. |
Delta |
vector of censoring indicators. |
X |
matrix of marginal mean covariates with one column for one covariate (design matrix excluding intercept). |
ID |
vector of cluster identifiers. |
Value
coef - estimate of coefficients.
exp(coef) - estimate of hazard ratio.
MDMR-var - MDMR bias-corrected sandwich variance estimate of coef.
References
Mancl, L. A., & DeRouen, T. A. (2001). A covariance estimator for GEE with improved smallāsample properties. Biometrics, 57(1), 126-134.
Wang, X., Turner, E. L., & Li, F. Improving sandwich variance estimation for marginal Cox analysis of cluster randomized trials. Under Review.
Examples
Y <- c(11,19,43,100,7,100,100,62,52,1,7,6)
Delta <- c(1,1,1,0,1,0,0,1,1,1,1,1)
X1 <- c(0,0,0,0,0,0,1,1,1,1,1,1)
X2 <- c(-19,6,-25,48,10,-25,15,22,17,-9,45,12)
ID <- c(1,1,2,2,3,3,4,4,5,5,6,6)
X <- X1
CoxBcv.mdmr(Y,Delta,X,ID)
X <- cbind(X1,X2)
CoxBcv.mdmr(Y,Delta,X,ID)