From 038dcddf85feab176f3414249bc39d63bdc29ee4 Mon Sep 17 00:00:00 2001
From: Daniel <mail@danielluedecke.de>
Date: Tue, 24 Oct 2023 23:15:24 +0200
Subject: [PATCH] some fixes

---
 R/binned_residuals.R    | 26 +++++++++++++++++++++-----
 man/binned_residuals.Rd | 12 ++++++++++++
 2 files changed, 33 insertions(+), 5 deletions(-)

diff --git a/R/binned_residuals.R b/R/binned_residuals.R
index 3210a228c..77b3bae78 100644
--- a/R/binned_residuals.R
+++ b/R/binned_residuals.R
@@ -11,6 +11,13 @@
 #' @param n_bins Numeric, the number of bins to divide the data. If
 #'   `n_bins = NULL`, the square root of the number of observations is
 #'   taken.
+#' @param ci Numeric, the confidence level for the error bounds.
+#' @param ci_type Character, the type of error bounds to calculate. Can be
+#'   `"gaussian"` (default), `"exact"` or `"boot"`.
+#' @param residuals Character, the type of residuals to calculate. Can be
+#'   `"response"` (default), `"pearson"` or `"deviance"`.
+#' @param iterations Integer, the number of iterations to use for the
+#'   bootstrap method. Only used if `ci_type = "boot"`.
 #' @param show_dots Logical, if `TRUE`, will show data points in the plot. Set
 #'   to `FALSE` for models with many observations, if generating the plot is too
 #'   time-consuming. By default, `show_dots = NULL`. In this case `binned_residuals()`
@@ -68,15 +75,18 @@ binned_residuals <- function(model,
                              show_dots = NULL,
                              ci = 0.95,
                              ci_type = c("gaussian", "exact", "boot"),
+                             residuals = c("response", "pearson", "deviance"),
                              iterations = 1000,
                              ...) {
-
+  # match arguments
   ci_type <- match.arg(ci_type)
-  fv <- stats::fitted(model)
+  residuals <- match.arg(residuals)
+
+  fitted_values <- stats::fitted(model)
   mf <- insight::get_data(model, verbose = FALSE)
 
   if (is.null(term)) {
-    pred <- fv
+    pred <- fitted_values
   } else {
     pred <- mf[[term]]
   }
@@ -88,7 +98,13 @@ binned_residuals <- function(model,
   }
 
   y0 <- .recode_to_zero(insight::get_response(model, verbose = FALSE))
-  y <- y0 - fv
+
+  # calculate residuals
+  y <- switch(residuals,
+    response = y0 - fitted_values,
+    pearson = .safe((y0 - fitted_values) / sqrt(fitted_values * (1 - fitted_values))),
+    deviance = .safe(stats::residuals(model, type = "deviance"))
+  )
 
   if (is.null(n_bins)) n_bins <- round(sqrt(length(pred)))
 
@@ -107,7 +123,7 @@ binned_residuals <- function(model,
 
     r <- switch(ci_type,
       gaussian = stats::qnorm(c((1 - ci) / 2, (1 + ci) / 2), mean = ybar, sd = sdev / sqrt(n)),
-      exact = stats:::binom.test(sum(y0[items]), n)$conf.int - fv,
+      exact = stats:::binom.test(sum(y0[items]), n)$conf.int,
       boot = .boot_binned_ci(y[items], ci, iterations)
     )
     names(r) <- c("CI_low", "CI_high")
diff --git a/man/binned_residuals.Rd b/man/binned_residuals.Rd
index f0362c091..b913ec53a 100644
--- a/man/binned_residuals.Rd
+++ b/man/binned_residuals.Rd
@@ -11,6 +11,7 @@ binned_residuals(
   show_dots = NULL,
   ci = 0.95,
   ci_type = c("gaussian", "exact", "boot"),
+  residuals = c("response", "pearson", "deviance"),
   iterations = 1000,
   ...
 )
@@ -33,6 +34,17 @@ time-consuming. By default, \code{show_dots = NULL}. In this case \code{binned_r
 tries to guess whether performance will be poor due to a very large model
 and thus automatically shows or hides dots.}
 
+\item{ci}{Numeric, the confidence level for the error bounds.}
+
+\item{ci_type}{Character, the type of error bounds to calculate. Can be
+\code{"gaussian"} (default), \code{"exact"} or \code{"boot"}.}
+
+\item{residuals}{Character, the type of residuals to calculate. Can be
+\code{"response"} (default), \code{"pearson"} or \code{"deviance"}.}
+
+\item{iterations}{Integer, the number of iterations to use for the
+bootstrap method. Only used if \code{ci_type = "boot"}.}
+
 \item{...}{Currently not used.}
 }
 \value{