SERVING THE QUANTITATIVE FINANCE COMMUNITY

Swedish
Topic Author
Posts: 4
Joined: September 24th, 2008, 9:22 am

### can i regress Garch risk?

if i have measused risk by Garch model, can i regress it in a linear function and measured it by OLS?many thanks

bostonquant
Posts: 108
Joined: March 26th, 2006, 7:46 pm

### can i regress Garch risk?

Look at the assumption of OLS. Quick answer. No.
Last edited by bostonquant on September 23rd, 2008, 10:00 pm, edited 1 time in total.

Swedish
Topic Author
Posts: 4
Joined: September 24th, 2008, 9:22 am

### can i regress Garch risk?

Thanks, but is there any method that i can use to measure the regression in which Garch rik is one of the variables?

Swedish
Topic Author
Posts: 4
Joined: September 24th, 2008, 9:22 am

### can i regress Garch risk?

Thanks, but is there any method that i can use to measure the regression in which Garch risk is one of the variables?

Siberian
Posts: 298
Joined: June 25th, 2003, 8:56 pm

### can i regress Garch risk?

imho there is nothing wrong with regressing r(t+1) on sigmahat(t+1), where the latter is out-of-sample forecast of the vol one time-period ahead, s.t. sigmahat(t+1) is not a function of r(t+1), on the surface it does not look like it will violate OLS assumptions;i think the sign of the coefficient will be the more important piece of info from that regression.
Last edited by Siberian on September 29th, 2008, 10:00 pm, edited 1 time in total.
 ABOUT WILMOTT

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...

 JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...

GZIP: On