Merge pull request #217 from Jacobluke-/paired-unequal-length

Fix a bug when receiving wide-format paired data with different group sizes.

Author: sangyu

.github/workflows/deploy.yaml

@@ -11,4 +11,4 @@ on:
 jobs:
   deploy:
     runs-on: ubuntu-latest
-    steps: [uses: fastai/workflows/quarto-ghp@master]
+    steps: [uses: fastai/workflows/quarto-ghp3@master]

.github/workflows/test-nbdev.yaml

@@ -4,4 +4,4 @@ on:  [workflow_dispatch, pull_request, push]
 jobs:
   test-nbdev:
     runs-on: ubuntu-latest
-    steps: [uses: fastai/workflows/nbdev-ci@master]
+    steps: [uses: fastai/workflows/nbdev3-ci@master]

README.md

@@ -173,6 +173,17 @@ snippets.
 
 ## How to cite
 
+**Getting over ANOVA: Estimation graphics for multi-group comparisons**
+
+*Zinan Lu, Jonathan Anns, Yishan Mai, Rou Zhang, Kahseng Lian, Nicole
+MynYi Lee, Shan Hashir, Lucas Wang Zhuoyu, A. Rosa Castillo Gonzalez,
+Joses Ho, Hyungwon Choi, Sangyu Xu, Adam Claridge-Chang*
+
+bioRxiv preprint 2026.
+[10.64898/2026.01.26.701654](http://dx.doi.org/10.64898/2026.01.26.701654)
+
+[PDF](https://www.biorxiv.org/content/10.64898/2026.01.26.701654v1.full.pdf)
+
 **Moving beyond P values: Everyday data analysis with estimation plots**
 
 *Joses Ho, Tayfun Tumkaya, Sameer Aryal, Hyungwon Choi, Adam

dabest/_api.py

@@ -2,10 +2,10 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/API/load.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['load', 'prop_dataset']
 
-# %% ../nbs/API/load.ipynb 4
+# %% ../nbs/API/load.ipynb #218e4f14
 def load(
     data,
     idx=None,
@@ -112,7 +112,7 @@ def load(
         ps_adjust,
     )
 
-# %% ../nbs/API/load.ipynb 5
+# %% ../nbs/API/load.ipynb #570ff65a
 import numpy as np
 from typing import Union, Optional
 import pandas as pd

dabest/_bootstrap_tools.py

@@ -1,9 +1,9 @@
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/API/bootstrap.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['bootstrap', 'jackknife_indexes', 'bca']
 
-# %% ../nbs/API/bootstrap.ipynb 3
+# %% ../nbs/API/bootstrap.ipynb #4231300f
 import numpy as np
 import pandas as pd
 import seaborn as sns
@@ -12,7 +12,7 @@
 from scipy.stats import mannwhitneyu, wilcoxon, norm
 import warnings
 
-# %% ../nbs/API/bootstrap.ipynb 4
+# %% ../nbs/API/bootstrap.ipynb #e86b4b8d
 class bootstrap:
     """
     Computes the summary statistic and a bootstrapped confidence interval.
@@ -230,7 +230,7 @@ def __repr__(self):
         b = "[{} CI: {}, {}]".format(self.ci, self.bca_ci_low, self.bca_ci_high)
         return "\n".join([a, b])
 
-# %% ../nbs/API/bootstrap.ipynb 5
+# %% ../nbs/API/bootstrap.ipynb #00c814b9
 def jackknife_indexes(data):
     # Taken without modification from scikits.bootstrap package.
     """

dabest/_dabest_object.py

@@ -2,10 +2,10 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/API/dabest_object.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['Dabest']
 
-# %% ../nbs/API/dabest_object.ipynb 5
+# %% ../nbs/API/dabest_object.ipynb #d3c6f47a
 # Import standard data science libraries
 import warnings
 from numpy import array, repeat, random, issubdtype, number
@@ -14,7 +14,7 @@
 from scipy.stats import norm
 from scipy.stats import randint
 
-# %% ../nbs/API/dabest_object.ipynb 7
+# %% ../nbs/API/dabest_object.ipynb #350b12c1
 class Dabest(object):
 
     """
@@ -559,14 +559,12 @@ def _check_errors(self, x, y, idx, experiment, experiment_label, x1_level):
         self.__x1_level = x1_level
 
         if self.__is_paired and self.__output_data.isnull().values.any():
-            warn1 = f"NaN values detected under paired setting and removed,"
+            warn1 = f"NaN values detected under paired setting,"
             warn2 = f" please check your data."
             warnings.warn(warn1 + warn2)
             if x is not None and y is not None:
                 rmname = self.__output_data[self.__output_data[y].isnull()][self.__id_col].tolist()
                 self.__output_data = self.__output_data[~self.__output_data[self.__id_col].isin(rmname)]
-            elif x is None and y is None:
-                self.__output_data.dropna(inplace=True)
 
         # Check if there is a typo on paired
         if self.__is_paired and self.__is_paired not in ("baseline", "sequential"):

dabest/_delta_objects.py

@@ -2,10 +2,10 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/API/delta_objects.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['DeltaDelta', 'MiniMetaDelta']
 
-# %% ../nbs/API/delta_objects.ipynb 5
+# %% ../nbs/API/delta_objects.ipynb #1814896d
 from scipy.stats import norm
 import pandas as pd
 import numpy as np
@@ -15,7 +15,7 @@
 import warnings
 import datetime as dt
 
-# %% ../nbs/API/delta_objects.ipynb 6
+# %% ../nbs/API/delta_objects.ipynb #1bb53e06
 class DeltaDelta(object):
     r"""
     A class to compute and store the delta-delta statistics for experiments with a 2-by-2 arrangement where two independent variables, A and B, each have two categorical values, 1 and 2. The data is divided into two pairs of two groups, and a primary delta is first calculated as the mean difference between each of the pairs:
@@ -382,7 +382,7 @@ def results(self):
             self.__compute_results()
         return self.__results
 
-# %% ../nbs/API/delta_objects.ipynb 10
+# %% ../nbs/API/delta_objects.ipynb #18462ec5
 class MiniMetaDelta(object):
     """
     A class to compute and store the weighted delta.

dabest/_effsize_objects.py

@@ -2,10 +2,10 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../nbs/API/effsize_objects.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['TwoGroupsEffectSize', 'EffectSizeDataFrame', 'PermutationTest']
 
-# %% ../nbs/API/effsize_objects.ipynb 5
+# %% ../nbs/API/effsize_objects.ipynb #430d4ea8
 import pandas as pd
 import lqrt
 from scipy.stats import norm
@@ -23,7 +23,7 @@
 from string import Template
 import scipy.stats as spstats
 
-# %% ../nbs/API/effsize_objects.ipynb 6
+# %% ../nbs/API/effsize_objects.ipynb #9d988bdd
 class TwoGroupsEffectSize(object):
 
     """
@@ -818,7 +818,7 @@ def bec_pct_high(self):
         return self.__bec_pct_high
 
 
-# %% ../nbs/API/effsize_objects.ipynb 10
+# %% ../nbs/API/effsize_objects.ipynb #024b1d00
 class EffectSizeDataFrame(object):
     """A class that generates and stores the results of bootstrapped effect
     sizes for several comparisons."""
@@ -1658,7 +1658,7 @@ def delta2(self):
     def is_delta_delta(self):
         return self.__delta2
 
-# %% ../nbs/API/effsize_objects.ipynb 29
+# %% ../nbs/API/effsize_objects.ipynb #5d49f77f
 class PermutationTest:
     """
     A class to compute and report permutation tests.

dabest/_stats_tools/confint_1group.py

@@ -2,17 +2,17 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../../nbs/API/confint_1group.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['create_bootstrap_indexes', 'compute_1group_jackknife', 'compute_1group_acceleration', 'compute_1group_bootstraps',
            'compute_1group_bias_correction', 'summary_ci_1group']
 
-# %% ../../nbs/API/confint_1group.ipynb 4
+# %% ../../nbs/API/confint_1group.ipynb #9181f236
 import numpy as np
 from numpy.random import PCG64, RandomState
 from scipy.stats import norm
 from numpy import sort as npsort
 
-# %% ../../nbs/API/confint_1group.ipynb 5
+# %% ../../nbs/API/confint_1group.ipynb #bac09924
 def create_bootstrap_indexes(array, resamples=5000, random_seed=12345):
     """Given an array-like, returns a generator of bootstrap indexes
     to be used for resampling.

dabest/_stats_tools/confint_2group_diff.py

@@ -2,13 +2,13 @@
 
 # AUTOGENERATED! DO NOT EDIT! File to edit: ../../nbs/API/confint_2group_diff.ipynb.
 
-# %% auto 0
+# %% auto #0
 __all__ = ['create_jackknife_indexes', 'create_repeated_indexes', 'compute_meandiff_jackknife', 'bootstrap_indices',
            'compute_bootstrapped_diff', 'delta2_bootstrap_loop', 'compute_delta2_bootstrapped_diff',
            'compute_meandiff_bias_correction', 'compute_interval_limits', 'calculate_group_var',
            'calculate_bootstraps_var', 'calculate_weighted_delta']
 
-# %% ../../nbs/API/confint_2group_diff.ipynb 4
+# %% ../../nbs/API/confint_2group_diff.ipynb #fa733643
 import numpy as np
 from numpy import arange, delete, errstate
 from numpy import mean as npmean
@@ -18,7 +18,7 @@
 from scipy.stats import norm
 from numpy import isnan
 
-# %% ../../nbs/API/confint_2group_diff.ipynb 5
+# %% ../../nbs/API/confint_2group_diff.ipynb #8cf9b1fc
 @njit(cache=True, parallel=True)
 def create_jackknife_indexes(data):
     """