1
$\begingroup$

I'm researching on stock returns and my final goal is to write around 70 pages on their distribution.

So far I have gathered data for several indices and companies using R and chosen some distributions to model my data with. For example; normal distribution, t distribution, skewed t, skewed normal, generalised lambda distribution etc...

The problem i'm having is justifying why I picked these distributions.

I thought of maybe plotting histograms of the best normal,t,lambda curve and then showing that the fit is good enough for us to use this model.

If that's true, how do I know if the curve is good enough to assume this.

And are there any other ways to justify choosing these distributions?

$\endgroup$

1 Answer 1

0
$\begingroup$

Answering the question of which model to use for a given problem is an area of statistics unto itself, which is called model selection. Methods for this include AIC, cross-validation, and Bayesian model selection. Most introductory statistics textbooks will include material on model selection.

$\endgroup$
1
  • $\begingroup$ @HumzaAli Sure. If I answered your question to your satisfaction, you can accept my answer by clicking on the check mark under the voting arrows. $\endgroup$ Feb 11, 2017 at 18:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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