Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I have two Gaussian distributions with mean and variance $(\mu_1,\sigma^2_1$) and $(\mu_2,\sigma^2_2)$. I then receive a series of values $x_1, x_2,...,x_n$ with mean and variance $(\mu_3,\sigma^2_3)$.

Assuming that there is equal chance that the the $x$'s were generated from the first or second Gaussians (and all of the $x$'s were generated from the same distribution), how can I determine - or even quantify - which of the two distributions is more likely to have been the source?

(In case I'm butchering terminology: I have a distribution for events of type A, and a distribution for events of type B, and I receive an unlabeled set of data - I want to determine whether it's more likely that it's data of events of type A or events of type B).

share|improve this question

2 Answers 2

If you believe ALL the new data come from exactly one of the two, you could try hypothesis testing. For individual data points, you might want to check out "Discriminant Analysis".

share|improve this answer

You could try to apply a Parzen-Rosenblatt window on your $(x)$ distribution and compare it to your source distribution.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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