inmoose.utils.lm_fit

class inmoose.utils.lm_fit(design, b)

Python equivalent to R lm.fit

This class is merely a wrapper around RegressionResults, the return type of OLS.fit() and WLS.fit().

It contains the solution to the equation \(A X = B\), where \(A\) is a known matrix, \(B\) a known vector and \(X\) the unknown to solve for. The solution is obtained by a least-square-error fit based on the QR-decomposition of \(A\).

Parameters:

• design (array_like) – matrix \(A\) of the equation

• b (array_like) – offset (vector \(B\) in the equation)

__init__(design, b)

Methods

__init__(design, b)

Attributes

coefficients	the value found for \(X\)
df_residuals	degrees of freedom of residuals
effects	vector of orthogonal single-df effects.
fitted_values	the value \(A X\)
qr	the QR decomposition of matrix \(A\)
rank	the column rank of matrix \(A\)
residuals	the residuals, i.e. the difference between \(A X\) and \(B\)
weights	the weights used in the least-square-error fit