Hypothesis Testing — Analysis Tool Box

Run the most common statistical tests with built-in assumption checking, effect size reporting, and interpretable output.

Highlight functions

`ConductLinearRegressionAnalysis`

OLS regression with full diagnostic plots — residuals, Q-Q plot, Cook's distance, and a coefficient table with confidence intervals.

from analysistoolbox.hypothesis_testing import ConductLinearRegressionAnalysis

results = ConductLinearRegressionAnalysis(
    dataframe=df,
    outcome_variable="sales",
    predictor_variables=["marketing_spend", "price", "seasonality"]
)

`OneSampleTTest`

Compare a sample mean to a known population value.

from analysistoolbox.hypothesis_testing import OneSampleTTest

OneSampleTTest(
    dataframe=df,
    column_name="response_time",
    population_mean=10.0
)

`ChiSquareTestOfIndependence`

Test whether two categorical variables are independent.

from analysistoolbox.hypothesis_testing import ChiSquareTestOfIndependence

ChiSquareTestOfIndependence(
    dataframe=df,
    categorical_variable_1="product_category",
    categorical_variable_2="purchase_made"
)

All functions

| Function | Description | |---|---| | ChiSquareTestOfIndependence | Test independence of two categorical variables | | ChiSquareTestOfIndependenceFromTable | Chi-square from a contingency table | | ConductCoxProportionalHazardRegression | Survival analysis | | ConductLinearRegressionAnalysis | OLS with full diagnostics | | ConductLogisticRegressionAnalysis | Binary outcome logistic regression | | OneSampleTTest | One-sample t-test | | OneWayANOVA | Compare means across groups | | TTestOfMeanFromStats | T-test from summary statistics | | TTestOfProportionFromStats | Proportion test from statistics | | TTestOfTwoMeansFromStats | Two-mean comparison from statistics | | TwoSampleTTestOfIndependence | Independent samples t-test | | TwoSampleTTestPaired | Paired samples t-test |