Skip to content

1009: don't round for PIT histograms #1010

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
seabbs merged 4 commits into from
Jun 24, 2025
Merged

1009: don't round for PIT histograms #1010

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
10d7ab7
don't round for PIT histograms
sbfnk Jun 24, 2025
300b8bd
add news item
sbfnk Jun 24, 2025
cd97abc
Merge branch 'main' into pit-dont-round
sbfnk Jun 24, 2025
7ede688
add PR number
sbfnk Jun 24, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions NEWS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@
- Added tolerance for numeric errors when checking that probabilities sum to one in ordinal forecasts (#997)
- Made computation of p-values optional in pairwise comparisons by allowing `test_type = NULL` in `compare_forecasts()`. When `test_type = NULL`, p-values will be `NA` (#978).
- Added a dependency on a scoringRules version >= 1.1.2 which is required for ordinal forecasts (#1006).
- A bug was fixed in PIT histograms from continuous values (#1010).

# scoringutils 2.1.0

Expand Down
3 changes: 1 addition & 2 deletions R/metrics-sample.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -549,7 +549,6 @@ pit_histogram_sample <- function(observed,
p_x <- rowSums(predicted <= observed) / n_pred

# PIT calculation is different for integer and continuous predictions
predicted <- round(predicted)
if (get_type(predicted) == "integer" && integers != "ignore") {
p_xm1 <- rowSums(predicted <= (observed - 1)) / n_pred
if (integers == "random") {
Expand All @@ -572,7 +571,7 @@ pit_histogram_sample <- function(observed,
pit_values <- p_x
}

if (integers != "nonrandom") {
if (get_type(predicted) != "integer" || integers != "nonrandom") {
pit_histogram <- hist(pit_values, breaks = quantiles, plot = FALSE)$density
}

Expand Down