lmreg (version 1.3)

hyptest: Test of a linear hypothesis in a linear model

Description

Carries out test of a single linear hypothesis in a linear model.

Usage

hyptest(lmobj, p, xi = 0, type = "both")

Value

Returns the estimated value of the linear combination of model parameters, its standard error, the t-statistic, the degrees of freedom and the p-value.

Arguments

lmobj

An object produced by lm fitting.

p

A numeric vector containing coefficients of the linear combination of model parameters.

xi

A numeric variable containing hypothesized value of the linear combination of model parameters (default = 0).

type

A character variable indicating the type of alternative: "upper" (one-sided), "lower" (one-sided) or "both" (default, two-sided).

Author

Debasis Sengupta <[e-mail]>, Jinwen Qiu <[e-mail]>

Details

It is assumed that all the model parameters are estimable and the linear model is homoscedastic and normal.

References

Sengupta and Jammalamadaka (2019), Linear Models and Regression with R: An Integrated Approach.

Examples

data(lifelength)
lmlife <- lm(Lifelength~factor(Category), data = lifelength)
p <- c(0,0,0,1,-1,0,0,0)
hyptest(lmlife, p, xi = 1, type = "upper")