0
$\begingroup$

I've backed out a time series of risk-neutral densities of GBP/USD options using a non-parametric approach in Matlab and would like to assess their forecast ability by applying the unconditional test in section 2.3 (page 11) of Christoffersen and Mazzotta "The Accuracy of Density Forecasts from Foreign Exchange Options", JFinEmetr (2005). However, I am not sure how to go about obtaining a series of probability transforms from truncated RNDs. I am also very new to Matlab so any guidance regarding the implementation of Christoffersen and Mazzotta (2005) would be greatly appreciated.

$\endgroup$
  • $\begingroup$ This is wierd,, the public (SSRN) version of Christoffen and Mazzotta does not have a section 5b. What kinds of "probability transforms" does it use? $\endgroup$ – noob2 Oct 20 '17 at 17:37
  • $\begingroup$ Sorry, it should be section 2.3 (page 11) in the SSRN version. I have edited my question to make the correction and have also provided a link to the paper. $\endgroup$ – user51462 Oct 20 '17 at 22:26
1
$\begingroup$

Calculate the cumulative RNDs - these form a bijection from the observed returns through to (0,1). Reverse the map so that a random uniform samples a return from the distribution. Non-parametric sampling that recovers the estimated terminal density goven by vanilla option prices.

$\endgroup$
  • $\begingroup$ Thanks for your answer James, I think it cleared up a lot. I am not sure I understand how the 'uniform' probability transform is related to the cdf, I saw this expression in a paper: $U_{t, h} \equiv \int_{-\infty}^{S_{t+h}}f_{t, h}(u)du\equiv F_{t, h}(S_{t+h})$. The right-hand side makes sense but I don't quite know how that is equivalent to the uniform probability transform, $U_{t, h}$. $\endgroup$ – user51462 Oct 21 '17 at 0:06
  • $\begingroup$ I was able to calculate the cdfs and have applied the inverse of the standard normal cdf to them to obtain a series of normal transform variables, would the next step be just applying the GMM to test whether their central moments are equal to those of the normal distribution? $\endgroup$ – user51462 Oct 21 '17 at 0:06

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.