Serving the Quantitative Finance Community

 
User avatar
chocolatemoney
Topic Author
Posts: 322
Joined: October 8th, 2008, 6:50 am

Hedge ratio with nonlinear cointegration

August 26th, 2013, 9:07 am

Hi,Let's say we have a nonlinear cointegrating relationship y = f(x) where x is a vector of independent tradable variables (f.ex shares).I am new to non-linear cointegration. How would you calculate the hedge ratios?I could just linearize f(x). Is there something I should take into account, on top of the numerical accuracy?Which methodology would you employ?Thanks!
 
User avatar
AnalyticalVega
Posts: 2260
Joined: January 16th, 2013, 5:03 am

Hedge ratio with nonlinear cointegration

August 27th, 2013, 4:28 pm

The whole idea of non-linear cointegration seems a bit silly. Yes markets are non-linear, but if your linear approx (i.e OLS). doesn't estimate well (and in many cases it won't work very well), why think a non-linear one will work better? There are an infinite number of different non-linearities..just saying..http://cowles.econ.yale.edu/P/cp/p13b/p1355.pdf
Last edited by AnalyticalVega on August 26th, 2013, 10:00 pm, edited 1 time in total.
 
User avatar
chocolatemoney
Topic Author
Posts: 322
Joined: October 8th, 2008, 6:50 am

Hedge ratio with nonlinear cointegration

August 28th, 2013, 5:28 am

Well, in my case, the choice of the functional form has a clear financial and economic meaning.Now it is time to numerically, locally linearize the outcome, so I can figure out the hedge ratios.Just trying to brainstorm if high school algebra does the job or if I am missing something..Anyway, thanks for the link.