SSRDecomp: Decompose Sums of Squares within a Linear Model

In ryan-heslin/RegLesson: Linear Regression from Scratch

Decompose Sums of Squares within a Linear Model

Description

Decompose Sums of Squares within a Linear Model

Usage

SSRDecomp(X, Y, reduced_betas, full_betas, intercept_fitted = TRUE)

Arguments

X	A design matrix for a linear model
Y	A response vector for a linear model
reduced_betas	Indices for beta coefficients, indexed by 0 (e.g., 0 for β_0, etc. ).
full_betas	Indices for coefficients of the full model. Defaults to all coefficients.
intercept_fitted	Logical. Does the model include an intercept? Default TRUE.

Details

This function decomposes the sums of squares within a reduced and full model and computes the full model's coefficient of partial determination: the portion of SSE of the reduced model it explains.

Value

Object of SSRDecomp class. It contains the SSR and SSE of the full and reduced models, along with the coefficient of partial correlation.

ryan-heslin/RegLesson documentation built on Aug. 5, 2022, 9:03 p.m.