The Review of Economics and Statistics · 2002 · 385 citations · 62 references
BusinessEconomicsFinanceEconometricsFinancial EconomicsTime Series AnalysisAsset Return ModelingBrownian MotionLong-memory ProcessesStochastic VolatilityVolatility AnalysisVolatility ModelingContinuous-time ModelsLong-memory Stochastic VolatilityMultifractal ProcessesSample Path PropertiesStandard Brownian MotionTime-varying Volatility DynamicsTime-varying Volatility ModelsVolatility
This paper investigates the multifractal model of asset returns (MMAR), a class of continuous-time processes that incorporate the thick tails and volatility persistence exhibited by many financial time series. The simplest version of the MMAR compounds a Brownian motion with a multifractal time-deformation. Prices follow a semi-martingale, which precludes arbitrage in a standard two-asset economy. Volatility has long memory, and the highest finite moments of returns can take any value greater than 2. The local variability of a sample path is highly heterogeneous and is usefully characterized by the local Hölder exponent at every instant. In contrast with earlier processes, this exponent takes a continuum of values in any time interval. The MMAR predicts that the moments of returns vary as a power law of the time horizon. We confirm this property for Deutsche mark/U.S. dollar exchange rates and several equity series. We develop an estimation procedure and infer a parsimonious generating mechanism for the exchange rate. In Monte Carlo simulations, the estimated multifractal process replicates the scaling properties of the data and compares favorably with some alternative specifications.
62
Generalized autoregressive conditional heteroskedasticity
Tim Bollerslev · Journal of Econometrics · 1986 · 22.3K citations · Full text
Business, Finance, Econometrics +11
Conditional Heteroskedasticity in Asset Returns: A New Approach
Daniel B. Nelson · Econometrica · 1991 · 10.4K citations