mqr.plot.anova.

model_means

mqr.plot.anova.model_means(result, response, factors, axs, ci_kws=None)

Plot the means of each combination of factor levels in an ANOVA result.

Parameters:

resultstatsmodels.regression.linear_model.RegressionResults

Result of calling fit on a statsmodel linear regression model.

responsestr

Name of response variable.

factorslist[str]

List of names of categorical factors.

axsarray_like[matplotlib.axes.Axes]

Axes for the plot. Must have the same number of elements as factors. Will be flattened before use.

ci_kws: dict, optional

Keyword args for confidence intervals. Passed to matplotlib..errorbar.

Examples

This example performs an ANOVA on sample data, then shows the means of the factors.

from statsmodels.formula.api import ols

data = pd.read_csv(mqr.sample_data('anova-glue.csv'), index_col='Run')
model = ols('adhesion_force ~ C(primer) + C(glue)', data=data)
result = model.fit()

fig, axs = plt.subplots(1, 2, figsize=(4, 2), layout='constrained')
mqr.plot.tools.sharey(fig, axs)
mqr.plot.anova.model_means(
    result,
    response='adhesion_force',
    factors=['primer', 'glue'],
    axs=axs)

(Source code, png, pdf)