/ Docs

Regression API

R-style formula regression with full diagnostics.

LinearRegressor

LinearRegressor()

OLS linear regression.

Methods

  • fit(y, X) → self
  • predict(X, interval='none', alpha=0.05) → DataFrame
  • summary() → str
  • diagnose() → str

RidgeRegressor

RidgeRegressor(alpha=1.0)

L2 regularization.

LassoRegressor

LassoRegressor(alpha=1.0)

L1 regularization (sparse solutions).

HuberRegressor

HuberRegressor(delta=1.345)

Robust regression — resistant to outliers.

QuantileRegressor

QuantileRegressor(quantile=0.5)

Median regression (quantile=0.5) or arbitrary quantiles.

RegressionDiagnostics

RegressionDiagnostics()

Methods

  • vif(X) → Variance Inflation Factors
  • shapiroWilk(residuals) → Normality test
  • breuschPagan(residuals, X) → Homoscedasticity test
  • durbinWatson(residuals) → Autocorrelation test
  • cooksDistance(y, X) → Influence analysis
  • fullReport(y, X, residuals) → Complete diagnostics string