Skip to content

Commit 67ccb62

Browse files
0.31.0
1 parent 4a49dfc commit 67ccb62

File tree

3 files changed

+13
-8
lines changed

3 files changed

+13
-8
lines changed

DESCRIPTION

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
Package: marginaleffects
22
Title: Predictions, Comparisons, Slopes, Marginal Means, and Hypothesis Tests
3-
Version: 0.30.0.7
3+
Version: 0.31.0
44
Authors@R:
55
c(person(given = "Vincent",
66
family = "Arel-Bundock",

NEWS.md

Lines changed: 4 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -1,17 +1,14 @@
11
# News {.unnumbered}
22

3-
## 0.30.0.7
3+
## 0.31.0
44

5-
New:
6-
7-
* `inferences(method = "conformal_full")` implements full conformal prediction for regression models. This provides distribution-free prediction intervals by refitting the model on augmented data for each test observation. Unlike split conformal or CV+, full conformal uses all training data without requiring a separate calibration set. It is computationally intensive but provides valid coverage guarantees. For tidymodels workflows, pass the training data explicitly via the `data_train` argument.
8-
9-
Breaking changes:
5+
Deprecated:
106

117
* The `conformal_calibration`, `conformal_train`, and `conformal_test` arguments in `inferences()` are deprecated (but kept for backward compatibility). Use the new `data_calib`, `data_train`, and `data_test`.
128

13-
## 0.30.1
9+
New:
1410

11+
* `inferences(method = "conformal_full")` implements full conformal prediction for regression models. This provides distribution-free prediction intervals by refitting the model on augmented data for each test observation. Unlike split conformal or CV+, full conformal uses all training data without requiring a separate calibration set. It is computationally intensive but provides valid coverage guarantees. For tidymodels workflows, pass the training data explicitly via the `data_train` argument.
1512
* Add support `MASS::lda()` models for estimates but not standard errors. Thanks to @friendly for feature request #1598.
1613

1714
Bugs:

inst/WORDLIST

Lines changed: 8 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -7,6 +7,7 @@ ARD
77
Aaditya
88
Adelie
99
Adelies
10+
Alessandro
1011
Anastasios
1112
Angelopoulos
1213
Arel
@@ -26,6 +27,7 @@ CRC
2627
Candes
2728
Carlisle
2829
Codecov
30+
Conformalized
2931
Customizable
3032
Demetri
3133
Diniz
@@ -40,6 +42,7 @@ Estimand
4042
Estimands
4143
Fabbri
4244
Foygel
45+
G'Sell
4346
GLMMs
4447
Gam
4548
Gentoos
@@ -61,6 +64,7 @@ J'VJ
6164
JAX
6265
JM
6366
JVJ
67+
Jing
6468
Jørgensen
6569
Kapre
6670
Kirill
@@ -97,6 +101,7 @@ Rdatasets
97101
Rdatatable
98102
Resul
99103
Rina
104+
Rinaldo
100105
Rmarkdown
101106
Robb
102107
Rohan
@@ -117,10 +122,12 @@ Tomz
117122
Typst
118123
Uffe
119124
Umit
125+
Wasserman
120126
Westfall
121127
Wittenberg
122128
XGBoost
123129
Xu
130+
Yaniv
124131
aje
125132
al
126133
arXiv
@@ -288,6 +295,7 @@ thornton
288295
threenum
289296
tibbles
290297
tidiers
298+
tidymodels
291299
timpipeseek
292300
tinytable
293301
tukey

0 commit comments

Comments
 (0)