Fits Cox model with treatment weights $$ w(A)= \sum_a I(A=a)/P(A=a|X) $$, computes standard errors via influence functions that are returned as the IID argument. Propensity scores are fitted using either logistic regression (glm) or the multinomial model (mlogit) when more than two categories for treatment. The treatment needs to be a factor and is identified on the rhs of the "treat.model".
phreg_IPTW(
formula,
data,
treat.model = NULL,
weight.var = NULL,
weights = NULL,
estpr = 1,
pi0 = 0.5,
...
)for phreg
data frame for risk averaging
propensity score model (binary or multinomial)
a 1/0 variable that indicates when propensity score is computed over time
may be given, and then uses weights*w(A) as the weights
to estimate propensity scores and get infuence function contribution to uncertainty
fixed simple weights
arguments for phreg call
Also works with cluster argument. Time-dependent propensity score weights can also be computed when weight.var is 1 and then at time of 2nd treatment (A_1) uses weights w_0(A_0) * w_1(A_1) where A_0 is first treatment.
data <- mets:::simLT(0.7,100,beta=0.3,betac=0,ce=1,betao=0.3)
dfactor(data) <- Z.f~Z
out <- phreg_IPTW(Surv(time,status)~Z.f,data=data,treat.model=Z.f~X)
summary(out)
#>
#> n events
#> 100 47
#> coeffients:
#> Estimate S.E. dU^-1/2 P-value
#> Z.f1 -0.24887 0.26446 0.20875 0.3467
#>
#> exp(coeffients):
#> Estimate 2.5% 97.5%
#> Z.f1 0.77968 0.46431 1.3093
#>
#>