Robust design reduces excessive process variation. Identification of dispersion effects plays an important role in it. Two main approaches are used to analyze the experiment where levels of noise factors are changed on purpose. Either the function of sample variance computed from replications over various levels of the noise factors, or response values themselves are modelled. The presence of heteroscedasticity may complicate the identification of dispersion effects. The aim of the simulation study described in the paper was to compare various ways of modelling heteroscedasticity and the efficiency of both main approaches. We confined ourselves to the assumption of normal random errors and used the t-test to identify the dispersion effects.
dispersion effects, heteroscedasticity, generalized linear model
In order to set up a list of libraries that you have access to,
you must first login
or sign up.
Then set up a personal list of libraries from your profile page by
clicking on your user name at the top right of any screen.