lm_fit: lm_fit

In leyaozh/lm.hw4: Linear Regression

Description

Fit a linear regression model.

Usage

lm_fit(x, y, add.intercept = FALSE, method = "qr", tol = 1e-07)

Arguments

x	a n * p matrix of predictor variables.
y	a length n vector(q = 1) or a n * q matrix of response variables.
add.intercept	logical. If TRUE, add intercept term to the linear regression model.
method	the method to fit the model(only method "qr" and "inv" is supported). Default is qr.
tol	the tolerance for detecting linear dependencies in the columns of x. Default is 1e-7.

Value

lm_fit returns a list containing following response values.

coefficients	q named vectors of coefficients.
residuals	q vectors of residuals, which equals response minus fitted values.
rank	the numeric rank of the fitted linear model.
fitted.values	q vectors of the fitted mean values.
df.residual	the degrees of freedom of residuals.
method	the method to be used for fitting the model.

The following response values are returned and printed only when method = "qr".

effects	q vectors of orthogonal single-df effects.
qr	the QR decomposition of matrix predictor variables.

The functions lm_summary and lm_anova are used to obtain a summary and analysis of variance table of the results.

Examples

n = 10; p = 5; q = 2;
x = matrix(rnorm(n * p), n, p) # no intercept
y1 = rnorm(n)
y2 = matrix(rnorm(n * q), n, q)

# no intercept, using method "qr"
z = lm_fit(x = x, y = y1)
# no intercept, using method "inv"
z = lm_fit(x = x, y = y1, method ="inv")
# with intercept, using method "qr"
z = lm_fit(x = x, y = y2, add.intercept = TRUE)
# with intercept, using method "inv"
z = lm_fit(x = x, y = y2, add.intercept = TRUE, method = "inv")

leyaozh/lm.hw4 documentation built on Dec. 3, 2019, 7:18 a.m.