Extractor functions, confidence intervals, residual plots and statistical tests for natural effect models.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
## S3 method for class 'neModel' coef(object, ...) ## S3 method for class 'neModelBoot' confint(object, parm, level = 0.95, type = "norm", ...) ## S3 method for class 'neModel' confint(object, parm, level = 0.95, ...) ## S3 method for class 'neModel' residualPlot(object, ...) ## S3 method for class 'neModel' residualPlots(object, ...) ## S3 method for class 'neModel' summary(object, ...) ## S3 method for class 'neModel' vcov(object, ...) ## S3 method for class 'neModel' weights(object, ...)
a fitted natural effect model object.
a specification of which parameters are to be given confidence intervals, either a vector of numbers or a vector of names. If missing, all parameters are considered.
the confidence level required.
the type of bootstrap intervals required. The default
confint yields bootstrap confidence intervals or confidence intervals based on the sandwich estimator (depending on the type of standard errors requested when fitting the
Bootstrap confidence intervals are internally called via the
boot.ci function from the boot package.
Confidence intervals based on the sandwich estimator are internally called via
The default confidence level specified in
level (which corresponds to the
conf argument in
boot.ci) is 0.95
and the default type of bootstrap confidence interval,
"norm", is based on the normal approximation.
Bias-corrected and accelerated (
"bca") bootstrap confidence intervals require a sufficiently large number of bootstrap replicates (for more details see
A summary table with large sample tests, similar to that for
glm output, can be obtained using
vcov returns either the bootstrap variance-covariance matrix (calculated from the bootstrap samples stored in
or the robust variance-covariance matrix (which is a diagonal block matrix of the original sandwich covariance matrix).
weights returns a vector containing the regression weights used to fit the natural effect model.
These weights can be based on
ratio-of-mediator probability (density) weights (only if the weighting-based approach is used)
inverse probability of treatment (exposure) weights (only if
xFit was specified in
residualPlots are convenience functions from the car package. These can be used to assess model adequacy.
For the bootstrap, z-values in the summary table are calculated by dividing the parameter estimate by its corresponding bootstrap standard error. Corresponding p-values in the summary table are indicative, since the null distribution for each statistic is assumed to be approximately standard normal. Therefore, whenever possible, it is recommended to focus mainly on bootstrap confidence intervals for inference, rather than the provided p-values.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
data(UPBdata) weightData <- neWeight(negaff ~ att + educ + gender + age, data = UPBdata) neMod <- neModel(UPB ~ att0 * att1 + educ + gender + age, family = binomial, expData = weightData, se = "robust") ## extract coefficients coef(neMod) ## extract variance-covariance matrix vcov(neMod) ## extract regression weights w <- weights(neMod) head(w) ## obtain bootstrap confidence intervals confint(neMod) confint(neMod, parm = c("att0")) confint(neMod, type = "perc", level = 0.90) ## summary table summary(neMod) ## residual plots library(car) residualPlots(neMod)