Return-based and Range-Based (Co)Variance Estimation - With an Application to Foreign Exchange Markets
Authors:
Preprint
- 1 March 2002
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
This paper analyzes and compares range-based and return-based variance/covariance estimates. We provide new results on the relative efficiency of the range. We show that the use of the range is compatible with time varying volatilities and we extend the range to a multivariate setup. A new estimator for the covariance, termed co-range, which is based on high and low prices, is proposed. The main properties of the new estimator are derived. Empirically, we find that range-based measures of variance, standard deviation, co-variance and correlation possess a much higher degree of persistence compared to conventional measures based on open/close prices.Keywords
This publication has 28 references indexed in Scilit:
- The Distribution of Realized Exchange Rate VolatilityJournal of the American Statistical Association, 2001
- High- and Low-Frequency Exchange Rate Volatility Dynamics: Range-Based Estimation of Stochastic Volatility Models2001
- Modeling and Forecasting Realized Volatility2001
- The Distribution of Stock Return Volatility2000
- Exchange Rate Returns Standardized by Realized Volatility are (Nearly) Gaussian2000
- Answering the Skeptics: Yes, Standard Volatility Models do Provide Accurate ForecastsInternational Economic Review, 1998
- Heterogeneous Information Arrivals and Return Volatility Dynamics: Uncovering the Long‐Run in High Frequency ReturnsThe Journal of Finance, 1997
- Intraday periodicity and volatility persistence in financial marketsJournal of Empirical Finance, 1997
- Fractionally integrated generalized autoregressive conditional heteroskedasticityJournal of Econometrics, 1996
- The Maximum Likelihood Estimation of Security Price Volatility: Theory, Evidence, and Application to Option PricingThe Journal of Business, 1984