# How to forecast volatility using high-frequency data?

There is a large literature covering volatility forecasts with high-frequency tick data. Much of this has surrounded the concept of "realized volatility", such as:

Other efforts have looked at high/low data to improve the forecast without including all the tick data.

Robert Almgrem has a nice lecture on the subject as part of his "Time Series Analysis and Statistical Arbitrage" course at NYU.

What's the best way for forecast volatility using high-frequency data?

Note: A similar question was previously asked on Wilmott.

-
Both PDFs link to the same place (and it's a temp.pdf file, so maybe UPenn just copies whatever file you choose to temp.pdf temporarily?). Off-the-cuff thought: model volatility per-trade, not per-time-unit and then model trades-per-time-unit separately. –  barrycarter Mar 18 '11 at 15:24
The link to the second paper should now be correct. –  Louis Marascio Sep 9 '11 at 13:24
The links to Almgrem's lecture notes and class at NYU are very nice. Some good reading in there. –  Louis Marascio Sep 9 '11 at 13:26

Relevant paper:

Efficient Estimation of Volatility using High Frequency Data (Zumbach, Corsi, and Trapletti 2002)