Regression Models

9 regression algorithms from classical OLS to constrained optimization, all with full statistical inference.

Categories

Linear Models

Classical least squares variants

OLSWLSRLS

Regularized Models

Penalty-based regression

RidgeElastic Net

Generalized Linear Models

Non-Gaussian response distributions

Poisson GLMALM

Constrained Models

Bounded and non-negative solutions

BLSNNLS

Common Output Structure

All regression functions return a STRUCT with these fields:

Field	Type	Description
coefficients	LIST(DOUBLE)	Regression coefficients
intercept	DOUBLE	Intercept (if fit_intercept=true)
std_errors	LIST(DOUBLE)	Standard errors
t_statistics	LIST(DOUBLE)	t-statistics
p_values	LIST(DOUBLE)	p-values
confidence_intervals	LIST(STRUCT)	Lower/upper bounds
r_squared	DOUBLE	R² (0-1)
adj_r_squared	DOUBLE	Adjusted R²
rmse	DOUBLE	Root mean squared error
aic	DOUBLE	Akaike Information Criterion
bic	DOUBLE	Bayesian Information Criterion
n	BIGINT	Number of observations
k	BIGINT	Number of predictors

All Functions

All CategoryConstrainedGeneralizedLinearRegularized

Function	Category	Description
anofox_stats_ols_fit_agg	Linear	Ordinary Least Squares regression
anofox_stats_wls_fit_agg	Linear	Weighted Least Squares regression
anofox_stats_rls_fit_agg	Linear	Recursive Least Squares regression
anofox_stats_ridge_fit_agg	Regularized	Ridge regression (L2 penalty)
anofox_stats_elastic_net_fit_agg	Regularized	Elastic Net (L1+L2 penalty)
anofox_stats_poisson_glm_fit_agg	Generalized	Poisson GLM for count data
anofox_stats_alm_fit_agg	Generalized	Augmented Linear Model
anofox_stats_bls_fit_agg	Constrained	Bounded Least Squares
anofox_stats_nnls_fit_agg	Constrained	Non-negative Least Squares

Showing 9 of 9