grasp.step {grasp}R Documentation

Internal GRASP function

Description

The function selects significant predictors for each response variable

Usage

grasp.step(gr.Yi)

Arguments

gr.Yi A vector containing the selected responses

Details

This function is the central part of GRASP. It selects significant predictors for each response variable. By default the stepwise procedure starts with a full model containing all potential variables (selX), and goes in loop to try to eliminate one variable at the time. At each step, the less significant variable is dropped from the model, and the loop starts again with the remaining variables. Tests are used to decide whether a variable should be dropped, kept or reintroduced. Available tests are : AIC, BIC, F, Chisq, BRUTO and CROSS

Author(s)

Anthony.Lehmann@unige.ch

See Also

grasp grasp, grasp.in grasp.in


[Package grasp version 2.5-7 Index]