PyMetaAnalysis 0.4.0__tar.gz → 0.5.0__tar.gz

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/.github/workflows/ci.yml

@@ -22,7 +22,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.10", "3.11", "3.12", "3.13"]
+        python-version: ["3.10", "3.11", "3.12", "3.13", "3.14"]
 
     steps:
       - name: Check out repository

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/.github/workflows/pages.yml

@@ -8,7 +8,7 @@ on:
 
 concurrency:
   group: pages
-  cancel-in-progress: true
+  cancel-in-progress: false
 
 jobs:
   build:
@@ -17,7 +17,6 @@ jobs:
     timeout-minutes: 10
     permissions:
       contents: read
-      pages: write
 
     steps:
       - name: Check out repository

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/.github/workflows/release.yml

@@ -35,14 +35,17 @@ jobs:
           cache: pip
           cache-dependency-path: pyproject.toml
 
-      - name: Install build tools
+      - name: Install release dependencies
         run: |
           python -m pip install --upgrade pip
-          python -m pip install build twine
+          python -m pip install ".[test]" build twine
 
       - name: Validate release metadata
         run: python tools/check_release.py --tag "$GITHUB_REF_NAME"
 
+      - name: Run tests with branch coverage
+        run: python -m pytest --cov=meta_analyze --cov-branch --cov-report=term-missing
+
       - name: Build distributions
         run: python -m build
 
@@ -91,7 +94,8 @@ jobs:
           path: dist/
 
       - name: Publish to PyPI
-        uses: pypa/gh-action-pypi-publish@release/v1
+        # release/v1.14, pinned because this step holds the PyPI OIDC permission.
+        uses: pypa/gh-action-pypi-publish@ba38be9e461d3875417946c167d0b5f3d385a247
 
   github-release:
     name: Create GitHub Release

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/CHANGELOG.md

@@ -6,6 +6,85 @@ Changes planned for the next release accumulate under `Unreleased`.
 
 ## Unreleased
 
+## 0.5.0 - 2026-07-25
+
+### Added
+
+- random-effects inverse-variance results now provide opt-in Q-profile
+  confidence intervals for tau-squared, tau, I-squared, and H-squared, with an
+  explicit formal-empty-set flag at the constrained `[0, 0]` boundary.
+
+### Changed
+
+- sensitivity and influence workflows now borrow internal fitted buffers during
+  refits instead of repeatedly materializing public defensive copies;
+- Meta-regression stores its classic coefficient covariance alongside the
+  selected inference covariance and reuses one shared precision-geometry
+  implementation across fitting and diagnostics.
+- independent `metafor` fixtures now cover categorical and multivariable
+  influence diagnostics, no-intercept Riley prediction intervals, and explicit
+  Mantel-Haenszel pooling correction; iterative failure paths have direct
+  regression tests.
+
+### Fixed
+
+- inverse-variance means, heterogeneity statistics, and pooling and
+  meta-regression tau-squared equations now use overflow-safe relative-weight
+  calculations at the supported float64 boundary;
+- subnormal variances and non-finite derived effects now raise explicit domain
+  errors instead of leaking runtime warnings or returning invalid results;
+- binary OR, RD, and Mantel-Haenszel arithmetic now avoids intermediate
+  overflow for very large finite counts;
+- the sensitivity guide no longer incorrectly states that Meta-regression
+  Cook's distance and DFBETAS are unavailable.
+- subgroup-differences tests now use classic model variances independently of
+  Hartung-Knapp confidence-interval adjustments;
+- leave-one-out and cumulative workflows now retain or skip, respectively,
+  reduced Mantel-Haenszel fits that are not estimable instead of aborting the
+  complete sensitivity analysis;
+- forest and subgroup-forest plots now reject non-positive displayed
+  coordinates before enabling a logarithmic axis;
+- uncorrected risk-ratio analyses now accept a zero non-event cell when the
+  study effect and sampling variance remain well defined.
+- Hartung-Knapp random-effects prediction intervals now use the selected
+  adjusted pooled-mean variance and are recorded as `HK-PR`, matching
+  `metafor` Riley predictions;
+- random-effects subgroup analyses now retain single-study subgroups through
+  an explicit, warned common-effect fallback instead of failing the complete
+  analysis;
+- tagged releases now rerun the full branch-coverage test suite before
+  distributions can be built and published.
+- tau-squared methods and SMD variance conventions now use `None` as the
+  context-sensitive default, so explicitly inapplicable settings raise domain
+  errors instead of being silently ignored;
+- duplicate study labels now add a row-position warning while preserving
+  `row_id` as the unique audit key;
+- report JSON now serializes `pd.NaT` study labels as `null` rather than the
+  string `"NaT"`.
+- Meta-regression with `missing="drop"` now determines complete-row exclusions
+  before validating moderator values, so invalid values in already excluded
+  rows cannot abort the analysis;
+- cumulative analysis now rejects ambiguous string `order` selectors that
+  exist in both source data and study results;
+- empty inputs now report that at least one study row is required, and binary
+  zero-cell errors identify when `correction_scope="none"` disables an
+  otherwise positive correction.
+- iterative tau-squared estimators now mark only an exact constrained zero as
+  a boundary solution, rather than treating every positive root below `atol`
+  as zero;
+- the exported pooling and Meta-regression tau-squared estimators now reject
+  insufficient study or residual degrees of freedom with domain-specific
+  errors;
+- the Mantel-Haenszel estimator now rejects empty and zero-total strata before
+  division, preventing NaN propagation and misleading variance diagnostics.
+- prediction-interval metadata is now `None` when too few studies prevent an
+  interval from being calculated;
+- categorical moderator encoding no longer conflates booleans, integers, and
+  floating-point values through Python's cross-type numeric equality;
+- CI now covers Python 3.14, Pages deployments are not cancelled mid-flight,
+  and the credential-bearing PyPI publisher action is pinned to an immutable
+  commit.
+
 ## 0.4.0 - 2026-07-23
 
 ### Added

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/CITATION.cff

@@ -8,8 +8,8 @@ authors:
   - family-names: Ding
     given-names: Zhaobo
     email: ding.zb@yahoo.com
-version: 0.4.0
-date-released: 2026-07-23
+version: 0.5.0
+date-released: 2026-07-25
 repository-code: https://github.com/ZhaoboDing/PyMetaAnalysis
 url: https://zhaoboding.github.io/PyMetaAnalysis/
 license: MIT

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: PyMetaAnalysis
-Version: 0.4.0
+Version: 0.5.0
 Summary: A pandas-first, auditable meta-analysis library for Python
 Project-URL: Documentation, https://zhaoboding.github.io/PyMetaAnalysis/
 Project-URL: Source, https://github.com/ZhaoboDing/PyMetaAnalysis
@@ -18,6 +18,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Medical Science Apps.
 Requires-Python: >=3.10
 Requires-Dist: numpy>=1.24
@@ -139,7 +140,10 @@ not individual-level or causal effects.
 Random-effects inverse-variance models support REML (default), Paule-Mandel,
 and DerSimonian-Laird tau-squared estimators. Mean confidence intervals support
 the normal default plus unmodified and safeguarded Hartung-Knapp variants.
-Eligible random-effects fits include an HTS prediction interval.
+Eligible random-effects fits include an HTS prediction interval under normal
+inference and an HK-PR interval under either Hartung-Knapp variant. Call
+`result.tau2_confidence_interval()` for a Q-profile interval around
+heterogeneity.
 
 Generic analyses accept exactly one of `variance=` or `standard_error=`.
 Standard errors are squared internally and the conversion is recorded in the
@@ -157,6 +161,7 @@ result.estimate
 result.display_estimate
 result.ci
 result.tau2
+tau2_interval = result.tau2_confidence_interval()
 result.i2
 result.i2_method
 result.diagnostics

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/README.md

@@ -90,7 +90,10 @@ not individual-level or causal effects.
 Random-effects inverse-variance models support REML (default), Paule-Mandel,
 and DerSimonian-Laird tau-squared estimators. Mean confidence intervals support
 the normal default plus unmodified and safeguarded Hartung-Knapp variants.
-Eligible random-effects fits include an HTS prediction interval.
+Eligible random-effects fits include an HTS prediction interval under normal
+inference and an HK-PR interval under either Hartung-Knapp variant. Call
+`result.tau2_confidence_interval()` for a Q-profile interval around
+heterogeneity.
 
 Generic analyses accept exactly one of `variance=` or `standard_error=`.
 Standard errors are squared internally and the conversion is recorded in the
@@ -108,6 +111,7 @@ result.estimate
 result.display_estimate
 result.ci
 result.tau2
+tau2_interval = result.tau2_confidence_interval()
 result.i2
 result.i2_method
 result.diagnostics

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/adr/0002-statistical-policy.md

@@ -2,6 +2,8 @@
 
 - Status: Accepted
 - Date: 2026-07-15
+- Amendment: the prediction-interval variance decision is superseded by
+  [ADR 0004](0004-hartung-knapp-prediction-intervals.md).
 
 ## Context
 
@@ -65,10 +67,11 @@ percentage in human-readable output.
 
 ### Prediction intervals
 
-Random-effects inverse-variance models use the Higgins-Thompson-Spiegelhalter
-prediction interval with `k - 2` degrees of freedom and the classic variance
-of the pooled mean. It is unavailable below three studies. With three or four
-studies it is calculated with an explicit uncertainty warning.
+The original decision used the Higgins-Thompson-Spiegelhalter prediction
+interval with `k - 2` degrees of freedom and the classic variance of the
+pooled mean for every confidence-interval method. ADR 0004 replaces that
+variance rule for Hartung-Knapp fits while retaining the study-count boundary
+and small-sample warning.
 
 ## Validation
 

pymetaanalysis-0.5.0/docs/adr/0004-hartung-knapp-prediction-intervals.md

@@ -0,0 +1,46 @@
+# ADR 0004: Hartung-Knapp prediction intervals
+
+- Status: Accepted
+- Date: 2026-07-24
+
+## Context
+
+ADR 0002 applied the classic pooled-mean variance to every `k - 2`
+random-effects prediction interval. That kept the interval independent of
+`ci_method`, but it diverged from `metafor` Riley predictions and the
+Hartung-Knapp Partlett-Riley option in R `meta`. Both use the covariance
+selected for mean inference inside the prediction variance.
+
+This difference was material when the unmodified Hartung-Knapp variance was
+below or above the classic variance. The result metadata still identified the
+interval as `HTS`, so callers could not discover the difference from method
+configuration alone.
+
+## Decision
+
+Random-effects inverse-variance prediction intervals retain the `k - 2`
+critical value and three-study minimum:
+
+```text
+mu_hat +/- t_(k - 2, 1 - alpha/2)
+          * sqrt(tau^2 + Var_selected(mu_hat))
+```
+
+- normal inference uses the classic pooled-mean variance and records `HTS`;
+- `hartung_knapp` uses its unmodified adjusted variance and records `HK-PR`;
+- `hartung_knapp_adhoc` uses its lower-bounded adjusted variance and records
+  `HK-PR`.
+
+Committed values are generated directly from
+`metafor::predict(fit, predtype="Riley")` for all three inference choices.
+Prediction intervals remain unavailable below three included studies and
+retain the explicit warning with three or four studies.
+
+## Consequences
+
+- prediction intervals and mean intervals use a coherent selected covariance;
+- HK and safeguarded HK prediction intervals can differ from the normal HTS
+  interval even when tau-squared and the pooled estimate are unchanged;
+- `result.method.prediction_interval_method` distinguishes `HTS` from
+  `HK-PR`;
+- this decision supersedes only the prediction-variance paragraph of ADR 0002.

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/input-data.md

@@ -54,7 +54,10 @@ zero.
 `row_id` is the zero-based input position. It is independent of the study
 label and remains stable through exclusion, subgroup, leave-one-out, and
 cumulative workflows. Use it when study labels are duplicated or not suitable
-as identifiers.
+as identifiers. Duplicate labels are accepted because several rows can
+legitimately originate from the same publication, but the result records the
+affected row positions in `warnings` so accidental duplication remains
+auditable.
 
 The complete row table is available from:
 
@@ -84,6 +87,13 @@ result = ma.meta_analysis(
 result.excluded_studies[["row_id", "study", "exclusion_reason"]]
 ```
 
+For Meta-regression, missingness is resolved across the outcome, uncertainty,
+study label, and every moderator before moderator values are validated.
+Consequently, a moderator value in an already excluded row is retained for
+audit but cannot abort or influence the fitted design matrix. Moderator values
+in included rows must still satisfy the declared numeric or categorical
+contract.
+
 Dropped rows do not contribute to pooled estimates, Q, tau-squared, prediction
 intervals, or weights. Missing subgroup labels are always rejected because
 silently assigning or dropping them would change the subgroup definition.
@@ -92,7 +102,7 @@ silently assigning or dropping them would change the subgroup definition.
 
 | Input family | Required validation |
 | --- | --- |
-| Generic | finite effect; finite, strictly positive sampling variance |
+| Generic | finite effect; finite, strictly positive sampling variance large enough for a finite float64 inverse weight |
 | Binary | integer event counts and totals; positive totals; `0 <= events <= total` |
 | Continuous | finite means/SDs; non-negative SDs; integer group sizes of at least 2 |
 
@@ -100,6 +110,12 @@ Binary and continuous APIs preserve their raw input columns in
 `study_results`. Derived effects, variances, correction indicators, and
 weights appear alongside them.
 
+All numerical calculations use float64. Internally derived effects and
+variances must remain finite, and every included sampling variance must be
+large enough to produce a finite inverse-variance weight. Values outside that
+representable range raise `InvalidStudyDataError` instead of returning
+non-finite fitted results.
+
 ## Exclusion is visible
 
 Rows excluded by a configured rule remain present with:

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/meta-regression.md

@@ -67,6 +67,10 @@ This produces terms such as `region[Asia]` and `region[North America]`, each
 relative to `Europe`. The reference never depends on row order. Undeclared
 levels, levels absent after exclusions, and string moderators omitted from
 `categorical=` are errors rather than implicit recoding decisions.
+Categorical matching preserves scalar kinds: Python and NumPy integers are
+equivalent, for example, but booleans do not match integer levels and
+floating-point values do not match integer levels. Declare levels using the
+same scalar kind as the observed moderator values.
 
 Formula parsing, automatic interactions, splines, and polynomial terms are not
 implemented. Construct those columns explicitly before fitting when they are

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/method-selection.md

@@ -17,6 +17,20 @@ variation is part of the estimand. A small number of studies makes
 between-study variance and interval estimation uncertain; it does not turn a
 random-effects question into a common-effect question.
 
+The high-level APIs intentionally have outcome-specific defaults:
+
+| API | Default model and pooling |
+| --- | --- |
+| `meta_analysis()` | random-effects inverse variance, REML |
+| `meta_continuous()` | random-effects inverse variance, REML |
+| `meta_binary()` | common-effect Mantel-Haenszel |
+| `meta_regression()` | mixed-effects inverse variance, REML |
+
+These defaults reflect common entry-point conventions, not an automatic model
+choice. Switching outcome APIs can therefore switch the model family; set
+`model` and, for binary data, `method` explicitly in analysis code and
+protocols.
+
 ## Pooling methods
 
 | Data and model | Available pooling method |
@@ -42,9 +56,17 @@ Random-effects inverse-variance models provide:
 
 REML and PM are iterative. Convergence, iteration count, and whether the
 solution reached the zero boundary are recorded in `result.diagnostics`.
+The boundary flag means the constrained estimate is exactly zero; a positive
+root remains a positive estimate even when it is smaller than the configured
+root-finding tolerance.
 Failure to converge raises `ConvergenceError`; it does not silently fall back
 to DL.
 
+The public default is `tau2_method=None`: it resolves to REML for a random- or
+mixed-effects fit. Common-effect and Mantel-Haenszel fits reject an explicitly
+supplied tau-squared method so analysis code cannot appear to request an
+estimator that was not used.
+
 ## Confidence intervals
 
 `ci_method="normal"` is the stable default and uses the classic normal
@@ -85,12 +107,36 @@ The definition is recorded as `result.i2_method`: `q_based` or
 `tau2_typical_variance`. Internally, I-squared remains a proportion from zero
 to one.
 
+## Heterogeneity confidence intervals
+
+A random-effects inverse-variance fit can invert generalized Q to quantify
+uncertainty around tau-squared:
+
+```python
+heterogeneity_interval = result.tau2_confidence_interval()
+heterogeneity_interval.ci
+heterogeneity_interval.i2_ci
+```
+
+The point estimate still follows the selected DL, PM, or REML estimator. The
+Q-profile interval is a separate procedure and therefore is identical across
+those point estimators for the same data and confidence level; it is not
+guaranteed to contain every possible point estimate.
+
+For unusually homogeneous data, both unconstrained bounds can lie below zero.
+The method then has a formal empty confidence set. PyMetaAnalysis exposes
+`is_empty=True` while representing the constrained tau-squared display as
+`[0, 0]`; do not interpret that display as proof that heterogeneity is exactly
+zero.
+
 ## Prediction intervals
 
-Random-effects models report an HTS prediction interval when at least three
-studies are included. It describes uncertainty for a new study's underlying
-effect, not uncertainty around the pooled mean. Common-effect models do not
-produce prediction intervals. With three or four studies the interval is still
+Random-effects models report a prediction interval when at least three studies
+are included. Under normal inference this is recorded as `HTS`; either
+Hartung-Knapp choice supplies its selected pooled-mean variance and is recorded
+as `HK-PR`. It describes uncertainty for a new study's underlying effect, not
+uncertainty around the pooled mean. Common-effect models do not produce
+prediction intervals. With three or four studies the interval is still
 calculated, but the result warns that it is especially uncertain.
 
 ## Subgroups
@@ -101,6 +147,11 @@ within each subgroup and again for the overall model. This is recorded as
 differences compares subgroup summary estimates; it is not a comparison of
 whether individual subgroup p-values are significant.
 
+If a subgroup contains only one included study, tau-squared and random-effects
+inference are not estimable within that subgroup. PyMetaAnalysis retains it as
+the study's common-effect estimate and normal interval, records the fallback
+in warnings, and continues the overall random-effects subgroup analysis.
+
 ## Reporting checklist
 
 At minimum, report:

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/r-interoperability.md

@@ -69,7 +69,7 @@ remain explicit attributes in Python.
 | `model="random"` | random-effects `rma.uni()` | `random=TRUE` | Requires a tau-squared policy |
 | `method="IV"` | inverse-variance weighting | `method="Inverse"` | Binary API only; generic and continuous fits are IV |
 | `method="MH"` | `rma.mh()` | `method="MH"` | Common-effect OR/RR only |
-| `tau2_method="REML"` | `method="REML"` | `method.tau="REML"` | PyMetaAnalysis random-effects default |
+| `tau2_method=None` (resolved as `"REML"`) | `method="REML"` | `method.tau="REML"` | PyMetaAnalysis random-effects default |
 | `tau2_method="PM"` | `method="PM"` | `method.tau="PM"` | Paule-Mandel |
 | `tau2_method="DL"` | `method="DL"` | `method.tau="DL"` | DerSimonian-Laird |
 
@@ -79,6 +79,13 @@ handling, or heterogeneity definitions. In particular, PyMetaAnalysis records
 tau-squared/typical-variance definition, while common-effect and MH results use
 the Q-based definition.
 
+`result.tau2_confidence_interval()` corresponds to
+`confint(fit, type="QP")` for an `rma.uni` random-effects fit. Both return
+Q-profile bounds for tau-squared and monotonic tau, I-squared, and H-squared
+transformations. PyMetaAnalysis additionally exposes `is_empty` so a formal
+empty confidence set is distinguishable from its constrained `[0, 0]`
+display.
+
 ## Confidence and prediction intervals
 
 | PyMetaAnalysis `ci_method` | `metafor` | R `meta` | Behavior |
@@ -87,9 +94,12 @@ the Q-based definition.
 | `"hartung_knapp"` | `test="knha"` | `method.random.ci="HK"` | Unmodified HK variance and t quantile |
 | `"hartung_knapp_adhoc"` | `test="adhoc"` | HK plus an explicitly selected ad hoc correction | HK variance cannot fall below the classic variance |
 
-Eligible random-effects fits include the documented HTS prediction interval.
-R packages offer additional prediction-interval choices, so matching the mean
-interval does not by itself guarantee a matching prediction interval.
+Eligible random-effects fits use the documented HTS prediction interval under
+normal inference. Either Hartung-Knapp choice instead uses its selected
+pooled-mean variance with `k-2` degrees of freedom (`HK-PR`), matching
+`predict(fit, predtype="Riley")` in `metafor`. R packages offer additional
+prediction-interval choices, so method settings must still be compared
+explicitly.
 
 Meta-regression uses a separate prediction rule. Its default corresponds to
 `predict(fit)` in `metafor`: normal inference uses a normal critical value and

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/sensitivity-analysis.md

@@ -18,13 +18,17 @@ print(influence.to_dataframe())
 The returned `LeaveOneOutResult` contains:
 
 - `original`, the fitted result supplied to the workflow;
-- `results`, one immutable refit per omitted included study;
+- `results`, one omission-aligned immutable refit or `None` per included study;
 - `table`, `summary()`, and `to_dataframe()`, which return defensive copies;
+- `failed`, the rows for reduced models that could not be estimated;
 - `warnings`, for workflow-level notes.
 
 The table identifies `omitted_row_id` and `omitted_study` and reports each
 refit's estimate, standard error, confidence interval, tau-squared, Q,
-I-squared, and H-squared.
+I-squared, and H-squared. It also records `refit_success`, `error_type`, and
+`error_message`. An unestimable deletion remains aligned in `results` as
+`None`, retains a table row with unavailable numeric values, and does not stop
+the remaining deletions.
 
 Originally excluded rows are never omission candidates. Common-effect analysis
 requires at least two included studies so each refit retains one. Random-effects
@@ -125,9 +129,17 @@ cumulative = result.cumulative(
 )
 ```
 
+A string selector must identify exactly one namespace. If the same column name
+exists in both the original source data and the result study table, it is
+ambiguous and raises `InvalidStudyDataError`. Pass the intended Series or array
+explicitly, for example `order=studies["publication_year"]`.
+
 `CumulativeMetaAnalysisResult.results` contains the estimable prefix fits, and
 `final` returns the last fit. Its table records the rows and study labels added
 at each step together with the same principal statistics as leave-one-out.
+If an otherwise eligible prefix is not estimable, it is skipped with an
+explicit warning. Its pending studies are included in the next estimable
+prefix row, so the cumulative path and final all-study fit remain complete.
 
 When multiple studies have the same order value, `collapse=True` adds the tied
 studies in one step:
@@ -184,7 +196,3 @@ design, outcome definition, risk of bias, and numerical leverage. They are not
 by themselves a reason to exclude the study. Likewise, a cumulative trend can
 describe the historical evidence path but does not remove time-related changes
 in methods, populations, or publication processes.
-
-The current Meta-regression workflow reports exact deleted-model fits and
-coefficient changes. It does not yet calculate Cook's distance, DFBETAS, or an
-automatic influential-study flag.

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/guides/zero-events.md

@@ -28,6 +28,11 @@ Available scopes are:
 Setting the correction to zero or the scope to `none` is rejected when it
 leaves an OR/RR undefined or an effect variance non-positive.
 
+For RR, a zero non-event cell is allowed when both arms still have positive
+event counts and the sampling variance is positive. For example, a study with
+events in every participant of one arm can be analyzed without correction.
+OR still requires all four cells to be positive.
+
 ## Double-zero and double-all studies
 
 A study with no events in either group, or events in every participant in both
@@ -47,8 +52,14 @@ rd_zero_variance = "correct"
 ```
 
 The RD itself remains the raw treatment risk minus control risk. Corrected
-counts are used only to form a positive sampling variance. To exclude all such
-studies before pooling, Q, tau-squared, and weight calculations, use:
+counts affect only its sampling variance. Under the default
+`correction_scope="only_zero_studies"`, the corrected variance is used for
+every retained RD table containing at least one zero cell, including
+single-zero tables whose uncorrected variance was already positive. The
+`rd_zero_variance` policy separately decides whether tables whose raw RD
+variance is exactly zero are retained or excluded. To exclude all such
+zero-variance studies before pooling, Q, tau-squared, and weight calculations,
+use:
 
 ```python
 rd_zero_variance = "exclude"

{pymetaanalysis-0.4.0 → pymetaanalysis-0.5.0}/docs/limitations.md

@@ -12,9 +12,14 @@ deferred functionality explicit.
 - common-effect and univariate random-effects inverse-variance models;
 - common-effect Mantel-Haenszel OR/RR;
 - DL, PM, and REML tau-squared estimators;
+- Q-profile confidence intervals for tau-squared and its monotonic tau,
+  I-squared, and H-squared transformations in random-effects inverse-variance
+  models;
 - normal and random-effects Hartung-Knapp mean intervals;
-- HTS prediction intervals;
-- independent subgroup fits and a formal subgroup-differences test;
+- HTS and Hartung-Knapp Partlett-Riley (`HK-PR`) prediction intervals;
+- independent subgroup fits and a formal subgroup-differences test, with a
+  warned common-effect representation when a random-effects subgroup contains
+  only one included study;
 - leave-one-out and cumulative repeated-fit workflows;
 - forest and descriptive funnel plots;
 - structured provenance and reports;
@@ -32,6 +37,7 @@ deferred functionality explicit.
 - Mantel-Haenszel risk differences or random-effects MH pooling;
 - formula parsing, automatic interactions/splines, stepwise moderator
   selection, automatic pairwise contrasts, or multiplicity correction;
+- Q-profile confidence intervals for residual tau-squared in Meta-regression;
 - multilevel, multivariate, network, dose-response, diagnostic-accuracy, or
   individual-participant-data meta-analysis;
 - robust variance estimation or dependent-effect clustering;