test: S3 generic method to do hypothesis tests in the mmi model and...

In JacobBergstedt/mmi: Multiple Model Inference

S3 generic method to do hypothesis tests in the mmi model and family objects.

Description

Generic method to do tests for the mmi model and family objects. For the models the appropriate tests will be done depending on model type. Currently, a t-test using sandwich estimated standard errors is done for mmi_lm and KRmodcomp Kenward-Rogers F-test is done for mmi_lmm objects. An lrt is done for all other model objects. The information is stored in a tibble. For a fam object, a test will be performed for each element in the family and the information will be binded together in a using a map-reduce pattern. The FDR-adjusted significance level is also calculated and added to the tibble.

Usage

test(object, ...)

## S3 method for class 'mmi_model'
test(object)

## S3 method for class 'mmi_lm'
test(object)

## S3 method for class 'mmi_lmm'
test(object)

## S3 method for class 'fam'
test(object)

Arguments

object

An mmi model object.

Methods (by class)

• mmi_model: Method for superclass linkS4class{mmi_model}

• mmi_lm: Performs a t-test with the sandwiched-estimated standard errors and sets up a tibble with the information.

• mmi_lmm: Performs an F test and setups a tibble with the information.

• fam: Maps over the family, performs a test for each model using the corresponding S3 method, constructs a tibble with the p-value of the test included, and then binds the tibbles together. Also computes the FDR-adjusted significance level.

JacobBergstedt/mmi documentation built on March 8, 2024, 8:25 a.m.