Regarding normality assumptions for model evaluation I have been told by my supervisor that it is not needed in the case of analyzing but is needed in forecasting only. i am looking for an explanation.

  • 4
    $\begingroup$ Hello and welcome ! More detail is needed. Please edit your question and describe your study design, the data you intend to collect and your research question. $\endgroup$ Nov 21, 2023 at 10:47
  • $\begingroup$ Also please describe your model (regression (and what kind)? Factor analysis? Time series?) And also what you think the assumptions are. $\endgroup$
    – Peter Flom
    Nov 21, 2023 at 10:51
  • 1
    $\begingroup$ i think you would be better off asking your supervisor what they meant $\endgroup$
    – seanv507
    Nov 21, 2023 at 11:31

1 Answer 1


From my understanding, more important than normally distributed variables (prior to modelling) is the normality of model residuals (evaluated from a Q-Q plot, for example) after your model has been fit. Is this relevant to your question? Here is a paper describing this misconception (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5436580/)

Also, more information on your problem would be helpful!


Not the answer you're looking for? Browse other questions tagged or ask your own question.