wilmott.com

This review of realized volatility is quite recent (2008) and complete.Realized Volatility: A review This article reviews the exciting and rapidly expanding literature on realized volatility. Afterpresenting a general univariate framework for estimating realized volatilities, a simple discretetime model is presented in order to motivate the main results. A continuous time specificationprovides the theoretical foundation for the main results in this literature. Cases with and withoutmicrostructure noise are considered, and it is shown how microstructure noise can cause severeproblems in terms of consistent estimation of the daily realized volatility. Independent anddependent noise processes are examined. The most important methods for providing consistentestimators are presented, and a critical exposition of different techniques is given. The finitesample properties are discussed in comparison with their asymptotic properties. A multivariatemodel is presented to discuss estimation of the realized covariances. Various issues relating tomodelling and forecasting realized volatilities are considered. The main empirical findings usingunivariate and multivariate methods are summarized.Would be nice to know what is the estimator you use to use realized volatility ? (the traditionnal close to close measure, parkinson, Garman&Klaas, Rogers& Satchell, GARCh and the hundreds more I'm missing)

Interesting, but not very useful from a market perspective of pricing derivatives on realized variance. In particular, i wonder how to build consistent arbitrage strategies with discrete realized variance as an asset. In practice, realized variance bundled in variance swap are quite liquid (although it depends on the underlying), and so we should be able to write an arbitrage strategy leading behind to a PDE. Unfortunately i don't find any paper dealing with this problem. I tried to solve it by myself but for the moment it's a failure. On the best attempt is the thesis of hans buehler on variance swap market model but he limits himself to continuous semi-martingale models...

Realized volatility is really sexy! Any ideas on some novel research topic? Also, it would be interesting to exchange ideas on our own research efforts

A bit more on realized volatilityEbens paper Meddahi paper