deprecate modified sl methods

benkeser · Jul 18, 2018 · 04801d8 · 04801d8
1 parent cbb9662
commit 04801d8
Show file tree

Hide file tree

Showing 5 changed files with 10 additions and 221 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -16,8 +16,6 @@ export(SL.npreg)
 export(adaptive_iptw)
 export(ci)
 export(drtmle)
-export(method.CC_LS_mod)
-export(method.CC_nloglik_mod)
 export(wald_test)
 importFrom(SuperLearner,SuperLearner)
 importFrom(SuperLearner,trimLogit)
@@ -45,7 +43,6 @@ importFrom(stats,binomial)
 importFrom(stats,cov)
 importFrom(stats,gaussian)
 importFrom(stats,glm)
-importFrom(stats,plogis)
 importFrom(stats,pnorm)
 importFrom(stats,predict)
 importFrom(stats,qnorm)
diff --git a/R/estimate.R b/R/estimate.R
@@ -93,7 +93,7 @@ estimateG <- function(A, W, DeltaY, DeltaA, SL_g, glm_g, a_0, tolg,
           Y = trainDeltaA,
           X = trainW, newX = validW, family = stats::binomial(),
           SL.library = SL_g$DeltaA, verbose = verbose,
-          method = "method.CC_nloglik_mod"
+          method = "method.CC_nloglik"
         )
         # get predicted probability of missing treatment
         gn_DeltaA <- fm_DeltaA$SL.predict
@@ -142,7 +142,7 @@ estimateG <- function(A, W, DeltaY, DeltaA, SL_g, glm_g, a_0, tolg,
           Y = as.numeric(trainA[trainDeltaA == 1] == a_0[1]),
           X = trainW[trainDeltaA == 1, , drop = FALSE], newX = validW,
           family = stats::binomial(), SL.library = SL_g$A,
-          verbose = verbose, method = "method.CC_nloglik_mod"
+          verbose = verbose, method = "method.CC_nloglik"
         )
         gn_A <- vector(mode = "list", length = 2)
         gn_A[[1]] <- fm_A$SL.predict
@@ -346,7 +346,7 @@ estimateG <- function(A, W, DeltaY, DeltaA, SL_g, glm_g, a_0, tolg,
               X = trainW[include & include2, , drop = FALSE],
               newX = validW, family = stats::binomial(),
               SL.library = SL_g$DeltaY, verbose = verbose,
-              method = "method.CC_nloglik_mod"
+              method = "method.CC_nloglik"
             )
             # name the fit
             name_DeltaY[a_ct] <- paste0(
@@ -365,7 +365,7 @@ estimateG <- function(A, W, DeltaY, DeltaA, SL_g, glm_g, a_0, tolg,
               A = trainA[include], trainW[include, , drop = FALSE]
             ),
             family = stats::binomial(), SL.library = SL_g$DeltaY,
-            verbose = verbose, method = "method.CC_nloglik_mod"
+            verbose = verbose, method = "method.CC_nloglik"
           )
 
           # get predictions back setting A = a for every a in a_0
@@ -610,7 +610,7 @@ estimateQ <- function(Y, A, W, DeltaA, DeltaY, SL_Q, glm_Q, a_0, stratify,
           X = data.frame(A = trainA, trainW)[include, , drop = FALSE],
           verbose = verbose, family = family, SL.library = SL_Q,
           method = ifelse(family$family == "binomial",
-            "method.CC_nloglik_mod", "method.CC_LS_mod"
+            "method.CC_nloglik", "method.CC_LS"
           )
         )
 
@@ -647,7 +647,7 @@ estimateQ <- function(Y, A, W, DeltaA, DeltaY, SL_Q, glm_Q, a_0, stratify,
             X = trainW[include2 & include, , drop = FALSE],
             newX = validW, verbose = verbose, family = family,
             SL.library = SL_Q, method = ifelse(family$family ==
-              "binomial", "method.CC_nloglik_mod", "method.CC_LS_mod")
+              "binomial", "method.CC_nloglik", "method.CC_LS")
           )
           list(est = fm$SL.predict, fm = fm)
         })
@@ -830,7 +830,7 @@ estimateQrn <- function(Y, A, W, DeltaA, DeltaY, Qn, gn, glm_Qr, SL_Qr,
                 trainDeltaA == 1 & trainDeltaY == 1]), newX = data.frame(
                 gn = valid_g
               ), family = family, SL.library = SL_Qr,
-              method = "method.CC_LS_mod"
+              method = "method.CC_LS"
             ))
             # if all weights = 0, use discrete SL
             if (!all(fm$coef == 0)) {
@@ -1020,15 +1020,15 @@ estimategrn <- function(Y, A, W, DeltaA, DeltaY, Qn, gn, SL_gr, tolg, glm_gr,
                 X = data.frame(Qn = train_Q),
                 newX = data.frame(Qn = valid_Q),
                 family = stats::gaussian(), SL.library = SL_gr,
-                method = "method.CC_LS_mod"
+                method = "method.CC_LS"
               )
               fm2 <- SuperLearner::SuperLearner(
                 Y = as.numeric(Aeqa &
                   trainDeltaA == 1 & trainDeltaY == 1),
                 X = data.frame(Qn = train_Q),
                 newX = data.frame(Qn = valid_Q),
                 family = stats::binomial(), SL.library = SL_gr,
-                method = "method.CC_nloglik_mod"
+                method = "method.CC_nloglik"
               )
               if (!all(fm1$coef == 0)) {
                 grn1 <- fm1$SL.predict
@@ -1049,7 +1049,7 @@ estimategrn <- function(Y, A, W, DeltaA, DeltaY, Qn, gn, SL_gr, tolg, glm_gr,
                 X = data.frame(Qn = train_Q, gn = train_g),
                 newX = data.frame(Qn = valid_Q, gn = valid_g),
                 family = stats::binomial(),
-                SL.library = SL_gr, method = "method.CC_nloglik_mod"
+                SL.library = SL_gr, method = "method.CC_nloglik"
               )
               if (!all(fm2$coef == 0)) {
                 grn2 <- fm2$SL.predict

diff --git a/R/method_fixes.R b/R/method_fixes.R
diff --git a/man/method.CC_LS_mod.Rd b/man/method.CC_LS_mod.Rd
diff --git a/man/method.CC_nloglik_mod.Rd b/man/method.CC_nloglik_mod.Rd