summary.tmleMSM: Summarization of the results of a call to the tmleMSM...

In tmle: Targeted Maximum Likelihood Estimation

Summarization of the results of a call to the tmleMSM function

Description

These functions are all methods for class tmleMSM, summary.tmleMSM objects

Usage

## S3 method for class 'tmleMSM'
summary(object, ...)
## S3 method for class 'tmleMSM'
print(x, ...)
## S3 method for class 'summary.tmleMSM'
print(x, ...)

Arguments

object	an object of class tmleMSM.
x	an object of class tmleMSM for summary functions, class summary.tmleMSM for print functions.
...	currently ignored.

Details

print.tmleMSM prints the estimate, standard error, p-value, and 95% confidence interval only. print.summary.tmleMSM, called indirectly by entering the command summary(result) (where result has class tmleMSM), outputs additional information.

Value

estimates	matrix of MSM parameter estimates, standard errors, pvalues, upper and lower bounds on 95% confidence intervals
sigma	variance-covariance matrix
Qmodel	working model used to obtain initial estimate of Q portion of the likelihood, if glm used
Qterms	terms in the model for Q
Qcoef	coefficient of each term in model for Q
gmodel	model used to estimate treatment mechanism g
gterms	terms in the treatment mechanism model
gcoef	coefficient of each term in model for treatment mechanism
gtype	description of estimation procedure for treatment mechanism, e.g. "SuperLearner"
g.AVmodel	model used to estimate h(A,V) (or h(A,T))
g.AVterms	terms in the model for h(A,V)
g.AVcoef	coefficient of each term in model for h(A,V)
g.AVtype	description of estimation procedure for h(A,V)
g.Deltamodel	model used to estimate missingness mechanism g.Delta
g.Deltaterms	terms in the missingness mechanism model
g.Deltacoef	coefficient of each term in model for missingness mechanism
g.Deltatype	description of estimation procedure for missingness
psi.Qinit	MSM parameter estimates based on initial (untargeted) estimated Q

Author(s)

Susan Gruber

See Also

tmleMSM

tmle documentation built on Aug. 23, 2023, 1:08 a.m.