lsRobTestMM: Test for Least Squares Bias Using Robust MM Regressions

In msalibian/RobStatTM: Robust Statistics: Theory and Methods

Test for Least Squares Bias Using Robust MM Regressions

Description

Test for Least Squares Bias Using Robust MM Regressions

Usage

lsRobTestMM(object, test = c("T", "T0"), ...)

Arguments

object	An MM regression fitted model whose class is *lmrobdetMM*.
test	A character vector indicating which of two type of tests "T" or "T0: are used, with type "T" the default.
...	Pass through parameters

Details

The original version of lsRobTestMM is the lsRobTest in the package *robust*. The function lsRobTest had options *T1* and *T2*. However, we only recommend *T2*, and deprecate *T1*. Accordingly we use *T* for the former *T2*, and use *T0* for the former *T1*, and we deprecate *T0*.

The *coefs* component of the list is a matrix whose row names are the names of the regression predictor variables, and whose columns *LS*, *Robust*, *Delta*, *Std.error*, *t-stat*, *p-value* contain respectively, the least squares and robust coefficient estimates, the differences in the coefficient estimates, the standard errors of the differences, the resulting t-statistic values, and the resulting z-test p-values.

The *full* component of the list is itself a list with components the full model quadratic form chi-squared statistic value (*stat*), the degrees of freedom (*df*), and the full model p value (*p.value*).

The *test* component of the list is a character value indicating which of the tests *T* and *T0* has been computed.

The *efficiency* component of the list is *NULL* when test *T* has been used, and is equal to the normal distribution efficiency of the *lmrobdetMM* estimate when test *T0* has been used.

Value

A list with component names coefs, full, test, efficiency

Author(s)

Kjell Konis

Examples

args(lsRobTestMM)

msalibian/RobStatTM documentation built on June 13, 2025, 12:45 p.m.