pandas - Python Pingouin Wech ANOVA: indexer is out-of-bounds

Question

I've got some troubles with the pingouin function welch_anova.

My test DataFrame:

         age  categ
0        44      0
1        21      2
...     ...    ...
336811   52      1
336812   30      0
336813   53      0
336814   27      2
336815   53      0

[336713 rows x 2 columns]

I run a ANOVA:

aov = pg.anova(data=df_test, dv="age", between="categ", detailed=True)
print(aov)

   Source            SS      DF            MS             F  p-unc       np2
0   categ  7.320873e+06       2  3.660437e+06  22711.788769    0.0  0.118868
1  Within  5.426722e+07  336710  1.611690e+02           NaN    NaN       NaN

Because my homoscedasticity test fail:

homog_var = pg.homoscedasticity(data=df_sales, dv='age', group='categ')
print(homog_var)

                   W  pval  equal_var
levene  13053.244984   0.0      False

I need to run a Welch_ANOVA:

aov = pg.welch_anova(data=df_test, dv='age', between='categ')
print(aov)

  Source  ddof1         ddof2           F  p-unc       np2
0  categ      2  49283.143894  34495.8422    0.0  0.118868

Wich seems ok but I got an index error in console:

IndexError: single positional indexer is out-of-bounds

I don't get this error without the welch_anova function

My DF is clean (no NAN values)

I've try with other DF and get the same issue.

Thanks in advance for any idea that could solve this.

question from:https://stackoverflow.com/questions/65880693/python-pingouin-wech-anova-indexer-is-out-of-bounds

Answer 1

Waitting for answers