In Wikipedia article (https://en.wikipedia.org/wiki/Nonparametric_statistics) about nonparametric statistics there is "Method of moments (statistics) with polynomial probability distributions" listed as non-parametric model.

How does it make sense? How method of estimation parameters can be non-parametric?

Parenthetically, is there any difference between "non-parametric" and "nonparametric"? Which is more common/correct?

  • 1
    $\begingroup$ In general, method of moments is not non-parametric. However, things like Generalized Methods of Moments are typically considered semi-parametric and only require a finite partition of the parameter space to be functionally specified. I think what the wiki page is referring to is something like this: journals.plos.org/plosone/article?id=10.1371/…. The idea is that we want to approximate an unknown pdf on some finite interval. We hypothesize a polynomial basis and use some empirical moment matching to make the approximation fit well. $\endgroup$ – Tyrel Stokes Jul 29 '20 at 18:04
  • $\begingroup$ Thanks, it sounds reasonably $\endgroup$ – Alexander Golys Jul 29 '20 at 18:35

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.