corybrunson · Oct 1, 2026 · Oct 2, 2026 · Oct 7, 2026
diff --git a/NEWS.md b/NEWS.md
@@ -42,13 +42,16 @@ This version introduces source code and unit tests for new engines and dials:
 
 Coordinated with [parsnip PR #1384](https://github.com/tidymodels/parsnip/pull/1384).
 
-### linear prediction type
+### new prediction types
 
-Linear predictions are enabled for the `clm`, `lrm`, `orm`, `vglm`, and `ordinalNet` ordinal regression engines and for the `vgam` generalized additive model engine.
+Linear predictions via `predict_linear_pred()` are enabled for the `clm`, `lrm`, `orm`, `vglm`, and `ordinalNet` ordinal regression engines and for the `vgam` generalized additive model engine.
 They consistently return a single column of linear predictors (without threshold contributions).
 
 Coordinated with [parsnip PR #1391](https://github.com/tidymodels/parsnip/pull/1391).
 
+Confidence interval predictions via `predict_confint()` are enabled for the `clm` engine.
+They return pairs of columns containing pointwise intervals for each class probability.
+
 ### threshold structure and parallel regression model arguments
 
 The `threshold_structure` model argument for `ordinal_reg()` controls what constraints, if any, are imposed on the ordered thresholds.

diff --git a/R/decision_tree-data.R b/R/decision_tree-data.R
@@ -12,7 +12,6 @@
 # `rpartScore::rpartScore` components
 
 make_decision_tree_rpartScore <- function() {
-
   parsnip::set_model_engine("decision_tree", "classification", "rpartScore")
   parsnip::set_dependency(
     "decision_tree",

diff --git a/R/decision_tree-rpartScore.R b/R/decision_tree-rpartScore.R
@@ -32,8 +32,11 @@ rpartScore_wrapper <- function(formula, data, ...) {
   data[[lhs]] <- as.integer(data[[lhs]])
   # execute call on modified inputs
   cl <- rlang::call2(
-    .fn = "rpartScore", .ns = "rpartScore",
-    formula = rlang::expr(formula), data = rlang::expr(data), ...
+    .fn = "rpartScore",
+    .ns = "rpartScore",
+    formula = rlang::expr(formula),
+    data = rlang::expr(data),
+    ...
   )
   rlang::eval_tidy(cl)
 }
diff --git a/R/gen_additive_mod-data.R b/R/gen_additive_mod-data.R
@@ -12,7 +12,6 @@
 # `VGAM::vgam` components
 
 make_gen_additive_mod_vgam <- function() {
-
   parsnip::set_model_engine("gen_additive_mod", "classification", "vgam")
   parsnip::set_dependency(
     "gen_additive_mod",
@@ -94,12 +93,11 @@ make_gen_additive_mod_vgam <- function() {
       pre = NULL,
       post = predict_VGAM_class_post,
       func = c(fun = "predict", pkg = "VGAM"),
-      args =
-        list(
-          object = quote(object$fit),
-          newdata = quote(new_data),
-          type = "response"
-        )
+      args = list(
+        object = quote(object$fit),
+        newdata = quote(new_data),
+        type = "response"
+      )
     )
   )
 
@@ -112,12 +110,11 @@ make_gen_additive_mod_vgam <- function() {
       pre = NULL,
       post = predict_VGAM_prob_post,
       func = c(fun = "predict", pkg = "VGAM"),
-      args =
-        list(
-          object = quote(object$fit),
-          newdata = quote(new_data),
-          type = "response"
-        )
+      args = list(
+        object = quote(object$fit),
+        newdata = quote(new_data),
+        type = "response"
+      )
     )
   )
 
@@ -133,15 +130,13 @@ make_gen_additive_mod_vgam <- function() {
         tibble::tibble(.pred_linear_pred = unname(beta_x))
       },
       func = c(fun = "predict", pkg = "VGAM"),
-      args =
-        list(
-          object = quote(object$fit),
-          newdata = quote(new_data),
-          type = "link"
-        )
+      args = list(
+        object = quote(object$fit),
+        newdata = quote(new_data),
+        type = "link"
+      )
     )
   )
-
 }
 
 # nocov end
diff --git a/R/ordinal_reg-VGAM.R b/R/ordinal_reg-VGAM.R
@@ -122,13 +122,14 @@
 #' }#VGAM>1.1-9
 #' @export
 VGAM_vglm_wrapper <- function(
-    formula, data,
-    family = "cumulative",
-    link = "logitlink",
-    parallel = FALSE,
-    Thresh = NULL,
-    ...,
-    call = rlang::caller_env()
+  formula,
+  data,
+  family = "cumulative",
+  link = "logitlink",
+  parallel = FALSE,
+  Thresh = NULL,
+  ...,
+  call = rlang::caller_env()
 ) {
   rlang::check_installed("VGAM")
   VGAM_1_1_9 <- utils::packageVersion("VGAM") == package_version("1.1.9")
@@ -149,18 +150,26 @@ VGAM_vglm_wrapper <- function(
   # execute nested call on modified inputs
   family_call <- if (VGAM_1_1_9) {
     rlang::call2(
-      .fn = family, .ns = "VGAM",
-      link = link, parallel = parallel, threshold = Thresh
+      .fn = family,
+      .ns = "VGAM",
+      link = link,
+      parallel = parallel,
+      threshold = Thresh
     )
   } else {
     rlang::call2(
-      .fn = family, .ns = "VGAM",
-      link = link, parallel = parallel, Thresh = Thresh
+      .fn = family,
+      .ns = "VGAM",
+      link = link,
+      parallel = parallel,
+      Thresh = Thresh
     )
   }
   cl <- rlang::call2(
-    .fn = "vglm", .ns = "VGAM",
-    formula = rlang::expr(formula), data = rlang::expr(data),
+    .fn = "vglm",
+    .ns = "VGAM",
+    formula = rlang::expr(formula),
+    data = rlang::expr(data),
     family = family_call,
     ...
   )
@@ -170,13 +179,14 @@ VGAM_vglm_wrapper <- function(
 #' @rdname VGAM_vglm_wrapper
 #' @export
 VGAM_vgam_wrapper <- function(
-    formula, data,
-    family = "cumulative",
-    link = "logitlink",
-    parallel = FALSE,
-    Thresh = NULL,
-    ...,
-    call = rlang::caller_env()
+  formula,
+  data,
+  family = "cumulative",
+  link = "logitlink",
+  parallel = FALSE,
+  Thresh = NULL,
+  ...,
+  call = rlang::caller_env()
 ) {
   rlang::check_installed("VGAM")
   VGAM_1_1_9 <- utils::packageVersion("VGAM") == package_version("1.1.9")
@@ -194,18 +204,26 @@ VGAM_vgam_wrapper <- function(
   # execute nested call on modified inputs
   family_call <- if (VGAM_1_1_9) {
     rlang::call2(
-      .fn = family, .ns = "VGAM",
-      link = link, parallel = parallel, threshold = Thresh
+      .fn = family,
+      .ns = "VGAM",
+      link = link,
+      parallel = parallel,
+      threshold = Thresh
     )
   } else {
     rlang::call2(
-      .fn = family, .ns = "VGAM",
-      link = link, parallel = parallel, Thresh = Thresh
+      .fn = family,
+      .ns = "VGAM",
+      link = link,
+      parallel = parallel,
+      Thresh = Thresh
     )
   }
   cl <- rlang::call2(
-    .fn = "vgam", .ns = "VGAM",
-    formula = rlang::expr(formula), data = rlang::expr(data),
+    .fn = "vgam",
+    .ns = "VGAM",
+    formula = rlang::expr(formula),
+    data = rlang::expr(data),
     family = family_call,
     ...
   )
@@ -256,17 +274,26 @@ match_ordinal_family <- function(family, call = rlang::caller_env()) {
 }
 
 match_ordinal_link_VGAM <- function(link, call = rlang::caller_env()) {
-  if (! is.character(link)) {
+  if (!is.character(link)) {
     return(link)
   }
   check_string(link, arg = "ordinal_link", call = call)
 
   if (
     # keep native values
-    ! link %in% c(
-      "logitlink", "probitlink", "logloglink", "clogloglink", "cauchitlink",
-      "foldsqrtlink", "logclink", "gordlink", "pordlink", "nbordlink"
-    )
+    !link %in%
+      c(
+        "logitlink",
+        "probitlink",
+        "logloglink",
+        "clogloglink",
+        "cauchitlink",
+        "foldsqrtlink",
+        "logclink",
+        "gordlink",
+        "pordlink",
+        "nbordlink"
+      )
   ) {
     # modify standardized values
     link <- rlang::arg_match0(
@@ -294,10 +321,11 @@ match_ordinal_link_VGAM <- function(link, call = rlang::caller_env()) {
 }
 
 match_threshold_structure_VGAM <- function(
-  Thresh, v_1_1_9 = FALSE,
+  Thresh,
+  v_1_1_9 = FALSE,
   call = rlang::caller_env()
 ) {
-  if (! is.character(Thresh)) {
+  if (!is.character(Thresh)) {
     return(Thresh)
   }
   check_string(Thresh, arg = "threshold_structure", call = call)
@@ -310,7 +338,7 @@ match_threshold_structure_VGAM <- function(
 
   if (
     # keep native values
-    ! Thresh %in% native_vals
+    !Thresh %in% native_vals
   ) {
     # modify standardized values
     Thresh <- rlang::arg_match0(

diff --git a/R/ordinal_reg-clm.R b/R/ordinal_reg-clm.R
@@ -60,13 +60,13 @@ clm_wrapper <- function(
   }
 
   args <- c(formulas, list(data = data))
-  if (! is.null(weights)) {
+  if (!is.null(weights)) {
     args$weights <- weights
   }
-  if (! is.null(link)) {
+  if (!is.null(link)) {
     args$link <- match_ordinal_link_clm(link, call = call)
   }
-  if (! is.null(threshold)) {
+  if (!is.null(threshold)) {
     threshold <- switch(
       threshold,
       flexible = "flexible",
@@ -85,18 +85,18 @@ clm_wrapper <- function(
   # printing the fit shows the formulas and engine arguments that were used.
   res$call <- rlang::call2(
     "clm",
-    !!! formulas,
+    !!!formulas,
     data = rlang::sym("data"),
-    !!! arg_exprs,
-    !!! dot_exprs,
+    !!!arg_exprs,
+    !!!dot_exprs,
     .ns = "ordinal"
   )
 
   res
 }
 
 match_ordinal_link_clm <- function(link, call = rlang::caller_env()) {
-  if (! is.character(link)) {
+  if (!is.character(link)) {
     return(link)
   }
   check_string(link, arg = "ordinal_link", call = call)
@@ -150,3 +150,24 @@ predict_clm_pre <- function(new_data, object) {
   }
   new_data
 }
+
+predict_clm_confint_post <- function(x, object) {
+  lvl <- object$lvl
+  bounds <- vector("list", 2L * length(lvl))
+  for (i in seq_along(lvl)) {
+    bounds[[2L * i - 1L]] <- unname(x$lwr[, i])
+    bounds[[2L * i]] <- unname(x$upr[, i])
+  }
+  names(bounds) <- c(
+    rbind(paste0(".pred_lower_", lvl), paste0(".pred_upper_", lvl))
+  )
+  res <- tibble::as_tibble(bounds)
+
+  if (isTRUE(object$spec$method$pred$conf_int$extras$std_error)) {
+    errors <- lapply(seq_along(lvl), function(i) unname(x$se.fit[, i]))
+    names(errors) <- paste0(".std_error_", lvl)
+    res <- tibble::as_tibble(c(bounds, errors))
+  }
+
+  res
+}