linear_regression

linear_regression(x, y, intercept=True)

Ordinary least squares by singular value decomposition.

Mirrors the C# LinearRegression class of the Numerics library: estimates Y = alpha + beta*X + e, e ~ N(0, sigma), via SVD.

Parameters

Name Type Description Default
x array_like A 2D array of predictors with one row per observation, or a 1D array for a single predictor. required
y array_like Responses, one per row of x. required
intercept bool Whether to fit an intercept. Default True. True

Returns

Name Type Description
LinearRegressionResult

Examples

>>> from corehydropy import linear_regression
>>> x = [[1, 2], [2, 1], [3, 4], [4, 3], [5, 5]]
>>> y = [3.1, 4.2, 8.1, 9.2, 13.0]
>>> fit = linear_regression(x, y)
>>> round(float(fit.r_squared), 6)
0.997992