search

ACCLAB / DABEST-python

Data Analysis with Bootstrapped ESTimation
https://acclab.github.io/DABEST-python/
Apache License 2.0
341 stars 47 forks source link

Statistical Question #47

Closed MCaviezel closed 5 years ago

MCaviezel commented 5 years ago

Hi, Thanks for this wonderful tool you provided!

I was wondering if it is statistically possible to include nuisance parameters in the group comparison? For example, if I want to compare a biomarker (DV) of two groups (IV1) and control for gender (IV2) and age (IV3). In a multiple regression setting, I would include these variables in the model or use them as random effects. (Biomarker ~ Group + Gender + Age, respectively Biomarker ~ Group + (1 | Gender) + (1 | Age)) Is this also possible in bootstrapping?

Thank you and best regards,

Marco

josesho commented 5 years ago

From what I can gather, you can do a bootstrap multiple regression, but this is currently out of the scope of DABEST.

If you're interested in performing an ANOVA-style 2x2 analysis, you can use DABEST to create the relevant plots; Rob Calin-Jagerman did a quick demo on Twitter.

Hope this helps!