1
$\begingroup$

I am now using Exponential distribution to model the the time intervals of a sequence of random events.Since I can choose several different lamdas for this model , I want to find out which lamda of them is fitting the training data best. I find Kullback–Leibler divergence from some paper is used to do this , but :

  • I don't know if this way is suitable

  • I don't know how to operate either

I am new to machine learning . Anyone can help me ?

$\endgroup$
1
$\begingroup$

it sounds like you just want to pick the lambda that fits the model best...as in estimate lambda (assuming the model is exponential). Using Kullback-Leibler you could assess the fit of the exponential distribution (with the appropriately estimated lambda) and compare it to say another distribution (e.g. gamma).

Of course if you are considering gamma you might as well just do this in the context of a hypothesis test because exponential is just a special case of gamma.

$\endgroup$

Your Answer

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

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