In R, I see support for ARCH models and for ARMA models (in the tseries package). How would we estimate the workhorse model where stock returns are ARMA with ARCH errors? I am aware of the paper by Andy Weiss. I have used this model quite a bit using stata and consider it a staple. I couldn't find mention of it in the tseries library. -- Ajay Shah Consultant ajayshah at mayin.org Department of Economic Affairs http://www.mayin.org/ajayshah Ministry of Finance, New Delhi