adaplmboxcox: Box-Cox-transformation and LRT for deviation form normality...

In schaarschmidt/mixADA: Normalization, mixture models and screening cutpoints for anti-drug-antibody reactions

Description

For internal use in the mixADA GUI! Simplistic grid search for Box-Cox-parameter lambda, and subsequent LRT for deviation from normality (lambda=1) and log-normality (lambda=0).

Usage

adaplmboxcox(resadapmixmod, normop,
 group = c("nonresponder", "responder", "all"))

Arguments

resadapmixmod	an object resulting fromfitting a mixture model xxx, as can be obtained by using
normop	character string naiming one of the options for normalization
group	a single character string, naming for which subset of biologoical samples to consider for the choice of the box-cox parameter of the data

Details

Uses only the subset specified in group,

backktransforms to origninal scale (if necessary due to normalization specified in normop)

performs grid search for box-cox lambda on the backtransformed data, resulting in a rough estimate for lambda

computes LRT for H0:lamda=1 and H0:lambda=0 based on this rough estimate

returns tables with estimates and corresponding liklihood as well as test results

Value

a list:

tabest	data.frame with 'estimates' resulting from the grid search
tabtest	data.fra,me with test results of the 'LRT'

further elements to be used as acptions in the GUI output ...

schaarschmidt/mixADA documentation built on May 29, 2019, 3:25 p.m.