Quick start

Installation

Install from PyPI:

pip install ritest-python

Then import:

from ritest import ritest

Basic use

Linear models

from ritest import ritest

res = ritest(
    df=df,
    permute_var="treat",
    formula="y ~ treat + x1 + x2",
    stat="treat",
    reps=1000,
)

Generic statistics

def stat_fn(d):
    return d.y.mean() - d.y[d.treat == 0].mean()

res = ritest(
    df=df,
    permute_var="treat",
    stat_fn=stat_fn,
    reps=1000,
)

Weights, strata, clusters

To specify designs, either for linear models or generic statistics, pass additional arguments:

res = ritest(
    df=df,
    permute_var="treat",
    formula="y ~ treat + x1",
    stat="treat",
    weights="w",
    strata="stratum_id",
    cluster="cluster_id",
)

Results

To see a summary of the results, use:

res.summary()

You can also get individual results, for example the \(p\)-value, using:

p_ri = res.pval

also works with: obs_stat, pval_ci, coef_ci_bounds, and coef_ci_band.