I have two regressions to perform - one with a metric DV (-3 to 3), the other with an ordered DV (0,1,2,3). Neither normal distribution nor homoscedasticity is given. I have a two questions:

  1. Some sources say robust regression take care of both lack of normal distribution and heteroscedasticity, while others say only of normal distribution. What is true?

  2. Are there ways of using robust regressions with ordered data, or is that only possible for metric DVs?

I have SPSS, R, and MPLUS available.

  • 1
    $\begingroup$ 1. It depends on the type of robust regression, I believe 2. It usually isn't necessary because ordinal logistic regression makes different sorts of assumptions (mostly about proportional odds). $\endgroup$
    – Peter Flom
    Oct 7 '12 at 23:37
  • 1
    $\begingroup$ 1. Is there a good overview about this topic for non-statisticians who want to apply these principles to data? Which robust regression can I use when both assumptions are not met? 2. What does "usually" mean? Googling "Ordered robust regression" literally finds nothing. Do you have a source for this topic? Thank you $\endgroup$
    – Torvon
    Oct 8 '12 at 22:16
  • 1
    $\begingroup$ Google "ordinal logistic regression" and you will get lots of stuff about it. One book I like is Long although this covers much more than ordinal logistic. Also good (but more techincal) is Agresti $\endgroup$
    – Peter Flom
    Oct 9 '12 at 0:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.