fastmatrix (version 0.5-7721)

ols.fit: Fitter functions for linear models

Description

This function is a switcher among various numerical fitting functions (ols.fit.cg, ols.fit.chol, ols.fit.qr, ols.fit.svd and ols.fit.sweep). The argument method does the switching: "qr" for ols.fit.qr, etc. This should usually not be used directly unless by experienced users.

Usage

ols.fit(x, y, method = "qr", tol = 1e-7, maxiter = 100)

Value

a list with components:

coefficients

a named vector of coefficients

residuals

the residuals, that is response minus fitted values.

fitted.values

the fitted mean values.

RSS

the residual sum of squares.

cov.unscaled

a \(p \times p\) matrix of (unscaled) covariances of the \(\hat\beta_j\), \(j=1, \dots, p\).

Arguments

x

design matrix of dimension \(n\times q\).

y

vector of observations of length \(n\).

method

currently, methods "cg", "chol", "qr" (default), "svd" and "sweep" are supported.

tol

tolerance for the conjugate gradients (gc) method. Default is tol = 1e-7.

maxiter

The maximum number of iterations for the conjugate gradients (gc) method. Defaults to 100.

See Also

ols.fit.cg, ols.fit.chol, ols.fit.qr, ols.fit.svd, ols.fit.sweep.

Examples

set.seed(151)
n <- 100
p <- 2
x <- matrix(rnorm(n * p), n, p) # no intercept!
y <- rnorm(n)
fm <- ols.fit(x = x, y = y, method = "chol")
fm