LSest {mffSM}R Documentation

Least squares estimates in a linear model

Description

It calculates the least squares estimates of the estimable parameters of a linear model including their standard errors and confidence intervals. It also performs the linear model based t-tests of hypotheses of equality of the estimable parameters to given values. Provided p-values and confidence intervals are calculated as if each parameter is estimated/tested separately from the others, i.e., no corrections for multiple testing are considered.

Usage

LSest(x, L, alternative = c("two.sided", "less", "greater"), conf.level = 0.95, theta0 = 0)

## S3 method for class 'LSest'
print(x, ...)

Arguments

x

an object of class lm for LSest, an object of class LSest for print.LSest.

L

either a numeric vector or a numeric matrix. If it is a vector, it provides the coefficients of the linear combination that determines the estimable parameter of the linear model to estimate and test. If it is a matrix then each row provides the coefficients of the linear combination that determines the estimable paramaters of the linear model to estimate and test.

alternative

a character string specifying the alternative hypothesis for all tests being performed, must be one of “two.sided” (default), “greater” or “less”. You can specify just the initial letter.

conf.level

confidence level of the intervals.

theta0

either a single number or a vector of length equal to the number of estimable parameters being estimated/tested. The single number is recycled into such a vector. It determines the values of each estimable parameter being estimated/tested under the null hypothesis.

...

additional arguments passed to the print method.

Value

A list with: estimates, standard errors, test statistics from the t-tests, p-values and lower and upper limit of the confidence intervals.

Author(s)

Arnošt Komárek arnost.komarek[AT]mff.cuni.cz

See Also

lm

Examples

### ANOVA model
data(Cars2004, package = "mffSM")
m1 <- lm(consumption ~ fdrive, data = Cars2004)
summary(m1)

### Estimate group differences for each pair
L <- matrix(c(0, 1, 0,  0, 0, 1,  0, -1, 1), ncol = 3, byrow = TRUE)
colnames(L) <- names(coef(m1))
rownames(L) <- c("rear-front", "4x4-front", "4x4-rear")
LSest(m1, L = L)

### One-sided tests with different null hypothesis value
### for each test
LSest(m1, L = L, alternative = "greater", theta0 = c(1, 2.5, 1))

[Package mffSM version 1.2 Index]