It is well known that the sample covariance matrix estimator results in a singular matrix if the variables is larger than the number of observations.

One solution is to apply shrinkage estimators. Are there functions/packages in mathematica that perform shrinkage estimation of the covariance matrix?

I am looking for something that is comparable to the R package corpcor.

  • $\begingroup$ Have you considered the possibility of using the corpcor package itself from within Mathematica through RLink? $\endgroup$ – MarcoB Jun 23 '15 at 18:10
  • $\begingroup$ I have not used RLink yes .. could give it a try but I would prefer something inside mathematica only. Thanks! $\endgroup$ – Richard Jun 24 '15 at 6:08
  • $\begingroup$ Richard, do you have any links you could suggest to the description of an algorithm that would do what you ask? $\endgroup$ – MarcoB Jun 25 '15 at 3:54

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.