Clear Filters
Clear Filters

Does this mean my dataset has too few observations?

9 views (last 30 days)
I ran a GLM regression with 5 indepedent variables. My dataset contains 238 observations and I got error message like this:
Warning: Regression design matrix is rank deficient to within machine precision. > In TermsRegression>TermsRegression.checkDesignRank at 98 In GeneralizedLinearModel>GeneralizedLinearModel.fit at 1174 In fitglm at 133
Does this mean I have too few observations? Because if I have 347 observations, it runs fine without any complaints.

Answers (1)

Star Strider
Star Strider on 17 Jul 2014
My guess is that there is something strange about Observation #238. Check it to see that all the data in it are valid.
  2 Comments
TingTing
TingTing on 17 Jul 2014
Hi, thanks a lot for reply! I wonder why you mentioned Observation #238. I plotted out leverage and ¤238 does not seem to be outlier. While #20 seems to be outlier. I excluded #20 and still got the same error message...
Star Strider
Star Strider on 17 Jul 2014
My point is that there is something about the added observation that is crashing your design matrix. I would check to see that all the data in that observation are valid.

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!