cogpy.regression.ols

Ordinary least-squares regression primitives.

Thin wrappers around numpy.linalg.lstsq for fitting, predicting, and computing residuals. Designed to work with design matrices from cogpy.regression.design.

Functions

ols_fit(X, Y, *[, rcond])	Fit ordinary least-squares: find beta minimizing ||Y - X @ beta||^2.
ols_predict(X, beta)	Compute predicted signal: Y_hat = X @ beta.
ols_residual(X, Y, beta)	Compute residual signal: Y - X @ beta.