Doc, namespace, and vignette updates

Author: ngreifer

DESCRIPTION

@@ -23,6 +23,7 @@ Imports:
     chk (>= 0.9.2),
     rlang (>= 1.1.0),
     crayon (>= 1.3.4),
+    sandwich,
     generics,
     utils,
     stats
@@ -41,7 +42,6 @@ Suggests:
     fwb (>= 0.2.0),
     splines,
     marginaleffects (>= 0.19.0),
-    sandwich,
     MASS,
     gbm (>= 2.1.9),
     dbarts (>= 0.9-29),

NAMESPACE

@@ -5,8 +5,6 @@ S3method(as.weightit,weightit.fit)
 S3method(as.weightitMSM,default)
 S3method(calibrate,default)
 S3method(calibrate,weightit)
-S3method(generics::estfun,multinom_weightit)
-S3method(generics::estfun,ordinal_weightit)
 S3method(generics::glance,coxph_weightit)
 S3method(generics::glance,glm_weightit)
 S3method(generics::glance,multinom_weightit)
@@ -32,6 +30,12 @@ S3method(print,summary.weightit.sbps)
 S3method(print,summary.weightitMSM)
 S3method(print,weightit)
 S3method(print,weightitMSM)
+S3method(sandwich::bread,glm_weightit)
+S3method(sandwich::bread,multinom_weightit)
+S3method(sandwich::bread,ordinal_weightit)
+S3method(sandwich::estfun,glm_weightit)
+S3method(sandwich::estfun,multinom_weightit)
+S3method(sandwich::estfun,ordinal_weightit)
 S3method(stats::anova,coxph_weightit)
 S3method(stats::anova,glm_weightit)
 S3method(stats::anova,multinom_weightit)
@@ -83,8 +87,9 @@ import(stats)
 importFrom(crayon,italic)
 importFrom(crayon,strikethrough)
 importFrom(crayon,underline)
-importFrom(generics,estfun)
 importFrom(generics,glance)
 importFrom(generics,tidy)
 importFrom(graphics,hist)
 importFrom(graphics,plot)
+importFrom(sandwich,bread)
+importFrom(sandwich,estfun)

man/glm_weightit-methods.Rd

@@ -19,7 +19,7 @@ editor_options:
 knitr::opts_chunk$set(echo = TRUE, eval=TRUE)
 options(width = 200, digits= 4)
 
-me_ok <- requireNamespace("marginaleffects", quietly = TRUE)
+me_ok <- requireNamespace("marginaleffects", quietly = TRUE) && !isTRUE(as.logical(Sys.getenv("NOT_CRAN", "false")))
 set.seed(1000)
 ```
 
@@ -97,7 +97,7 @@ fit <- lm_weightit(re78 ~ treat * (age + educ + race + married +
 # G-computation for the treatment effect
 library("marginaleffects")
 avg_comparisons(fit, variables = "treat",
-                newdata = subset(lalonde, treat == 1))
+                newdata = subset(treat == 1))
 ```
 
 Our confidence interval for `treat` contains 0, so there isn't evidence that `treat` has an effect on `re78`. Several types of standard errors are available in `WeightIt`, including analytical standard errors that account for estimation of the weights using M-estimation, robust standard errors that treat the weights as fixed, and bootstrapping. All type are described in detail at `vignette("estimating-effects")`.
@@ -141,7 +141,7 @@ Wmsm.out <- weightitMSM(list(A_1 ~ X1_0 + X2_0,
 Wmsm.out
 ```
 
-No matter which method is selected, `weightitMSM()` estimates separate weights for each time period and then takes the product of the weights for each individual to arrive at the final estimated weights. Printing the output of `weightitMSM()` provides some details about the function call and the output. We can take a look at the quality of the weights with `summary()`, just as we could for point treatments.
+`weightitMSM()` estimates separate weights for each time period and then takes the product of the weights for each individual to arrive at the final estimated weights. Printing the output of `weightitMSM()` provides some details about the function call and the output. We can take a look at the quality of the weights with `summary()`, just as we could for point treatments.
 
 ```{r}
 summary(Wmsm.out)
@@ -171,11 +171,10 @@ Then, we compute the average expected potential outcomes under each treatment re
 ```{r, eval = me_ok}
 library("marginaleffects")
 (p <- avg_predictions(fit,
-                      variables = c("A_1", "A_2", "A_3"),
-                      type = "response"))
+                      variables = c("A_1", "A_2", "A_3")))
 ```
 
-We can compare the expected potential outcomes under each regime using `marginaleffects::hypotheses()`. To get all pairwise comparisons, supply the `avg_predictions()` output to `hypotheses(., "pairwise")`. To compare individual regimes, we can use `hypotheses()`, identifying the rows of the `avg_predictions()` output. For example, to compare the regimes with no treatment for all three time points vs. the regime with treatment for all three time points, we would run
+We can compare the expected potential outcomes under each regime using `marginaleffects::hypotheses()`. To get all pairwise comparisons, supply the `avg_predictions()` output to `hypotheses(., ~ pairwise)`. To compare individual regimes, we can use `hypotheses()`, identifying the rows of the `avg_predictions()` output. For example, to compare the regimes with no treatment for all three time points vs. the regime with treatment for all three time points, we would run
 
 ```{r, eval = me_ok}
 hypotheses(p, "b8 - b1 = 0")

vignettes/WeightIt.Rmd

@@ -31,7 +31,7 @@ pre code {
 knitr::opts_chunk$set(echo = TRUE, eval = TRUE)
 options(width = 200, digits= 4)
 
-me_ok <- requireNamespace("marginaleffects", quietly = TRUE)
+me_ok <- requireNamespace("marginaleffects", quietly = TRUE) && !isTRUE(as.logical(Sys.getenv("NOT_CRAN", "false")))
 me_ok_new <- me_ok && utils::packageVersion("marginaleffects") > "0.25.0.1"
 su_ok <- requireNamespace("survival", quietly = TRUE)
 ```
@@ -582,12 +582,12 @@ hypotheses(a, ~pairwise)
 
 Though the subgroup effects differ from each other in the sample, this difference is not statistically significant at the .05 level, so there is no evidence of moderation by `X5`.
 
-When the moderator has more than two levels, it is possible to run an omnibus test for moderation by changing `hypothesis` to `"reference"` and supplying the output to `hypotheses()` with `joint = TRUE`, e.g.,
+When the moderator has more than two levels, it is possible to run an omnibus test for moderation by changing `hypothesis` to `~reference` and supplying the output to `hypotheses()` with `joint = TRUE`, e.g.,
 
 ```{r, eval=FALSE}
 avg_comparisons(fit, variables = "A",
                 by = "X5",
-                hypothesis = ~pairwise) |>
+                hypothesis = ~reference) |>
   hypotheses(joint = TRUE)
 ```