ENH: Add optional parameters for summary_col to indicate FEs

[adrianmross]

Researchers tend to show they used categorical variables without all the coefficients, saying whether they included a fixed effect. This PR adds that feature.

Example

import numpy as np
import pandas as pd
import seaborn as sns
import statsmodels.formula.api as smf
from statsmodels.iolib.summary2 import summary_col

# Load the diamonds dataset
diamonds = sns.load_dataset('diamonds')

# Adapted regressions for the diamonds dataset
regressions = [
    (smf.ols('np.log(price) ~ carat', data=diamonds).fit(), 'log(Price) ~ Carat'),
    (smf.ols('np.log(price) ~ np.log(carat)', data=diamonds).fit(), 'log(Price) ~ log(Carat)'),
    (smf.ols('np.log(price) ~ C(cut)', data=diamonds).fit(), 'log(Price) ~ C(Cut)'),
    (smf.ols('np.log(price) ~ C(clarity)', data=diamonds).fit(), 'log(Price) ~ C(Clarity)'),
    (smf.ols('np.log(price) ~ carat + C(cut) + C(clarity)', data=diamonds).fit(), 'log(Price) ~ Carat + C(Cut) + C(Clarity)')
]

# Output summary of all regressions
summary = summary_col([reg[0] for reg in regressions],
                      model_names=[f'{i}. '+reg[1] for i, reg in enumerate(regressions, 1)]
                    )

summary

With more complicated regressions or broader categories, this can quickly overwhelm the summary.

Now, with amended function.

"""
fixed_effects : list[str], optional
        List of categorical variables for which to indicate presence of fixed effects.
fe_present : str, optional
    String to indicate the presence of fixed effects. Default is "Yes".
fe_absent : str, optional
    String to indicate the absence of fixed effects. Default is empty string.
"""

summary = summary_col([reg[0] for reg in regressions],
                      model_names=[f'{i}. '+reg[1] for i, reg in enumerate(regressions, 1)],
                      fixed_effects=['cut', 'clarity'],
                    )

Much more concise.

[donbowen]

This looks like a nice feature!

[pep8speaks]

Hello @adrianmross! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2024-04-10 17:09:53 UTC