MATLAB Answers


Goodness of fit - Diebold, Gunther and Tay Approach - Probability Integral Transform - Kolmogorov

Asked by Johann
on 19 Jul 2013
Latest activity Answered by Di Lu
on 25 May 2014

Hi, folks!

I need to evaluate the goodness of fit of a few simple models for financial data. So far i have estimated the parameters for the following: GBM, Vasicek, GBM-Jump Diffusion, Heston-Nandi-GARCH and a two-state Markov chain regime switching model.

Now i want to know how well each of these models fit the data. What I have learned so far is that in order to perform the e.g. Kolmogorov test I need to apply first the probability integral transform by Diebold Gunther an Tay.

Does anyone of you know this method and can explain what exactly these guys propose for transformation? I would be grateful for any article where this transformation is explained step by step. (I've read quite a few on this but I still dont get it). Maybe anyone can provide some code?

Cheers Johann


1 Answer

Answer by Di Lu
on 25 May 2014

I also pay attention to the Markov Regime Switching model such as MRS-BEKK, but I find it is really hard to estimate. Therefor it is interesting how you solve the problem.


Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi test

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

MATLAB Academy

New to MATLAB?

Learn MATLAB today!