G-estimator for binomial regression model (Standardized estimates)

Computes G-estimator $$ \hat F(t,A=a) = n^{-1} \sum_i \hat F(t,A=a,Z_i) $$. Assumes that the first covariate is $A$. Gives influence functions of these risk estimates and SE's are based on these. If first covariate is a factor then all contrast are computed, and if continuous then considered covariate values are given by Avalues.

Usage

binregG(x, data, Avalues = NULL, varname = NULL)

Arguments

x

binreg object

data

data frame for risk averaging

Avalues

values to compare for first covariate A, assumes that first variable is factor and take all levels

varname

if given then averages for this variable, default is first variable

References

Blanche PF, Holt A, Scheike T (2022). “On logistic regression with right censored data, with or without competing risks, and its use for estimating treatment effects.” Lifetime data analysis, 29, 441–482.

Author

Thomas Scheike

Examples

library(mets)
data(bmt); bmt$time <- bmt$time+runif(408)*0.001
bmt$event <- (bmt$cause!=0)*1

b1 <- binreg(Event(time,cause)~age+tcell+platelet,bmt,cause=1,time=50)
sb1 <- binregG(b1,bmt,Avalues=c(0,1,2))
summary(sb1)
#> G-estimator :
#>       Estimate Std.Err   2.5%  97.5%   P-value
#> risk0   0.4058 0.02588 0.3551 0.4565 1.982e-55
#> risk1   0.5119 0.03706 0.4393 0.5846 2.057e-43
#> risk2   0.6168 0.05516 0.5087 0.7250 4.993e-29
#> 
#> Average Treatment effect: difference (G-estimator) :
#>      Estimate Std.Err    2.5%  97.5%   P-value
#> pa     0.1061 0.02623 0.05471 0.1575 5.222e-05
#> pa.1   0.2110 0.04960 0.11381 0.3082 2.096e-05
#> 
#> Average Treatment effect: ratio (G-estimator) :
#> log-ratio: 
#>       Estimate    Std.Err      2.5%     97.5%      P-value
#> [pa] 0.2323087 0.05277448 0.1288726 0.3357448 1.073002e-05
#> [pa] 0.4187166 0.08402886 0.2540231 0.5834101 6.260295e-07
#> ratio: 
#>      Estimate     2.5%    97.5%
#> [pa] 1.261509 1.137545 1.398982
#> [pa] 1.520010 1.289202 1.792139
#>