Estimation of time-dependent Hurst exponents with variational smoothing and application to forecasting foreign exchange rates

Hurst exponents depict the long memory of a time series. For human-dependent phenomena, as in finance, this feature may vary in the time. It justifies modelling dynamics by multifractional Brownian motions, which are consistent with time-varying Hurst exponents. We improve the existing literature on estimating time-dependent Hurst exponents by proposing a smooth estimate obtained by variational calculus. This method is very general and not restricted to the sole Hurst framework. It is globally more accurate and easier than other existing non-parametric estimation techniques. Besides, in the field of Hurst exponents, it makes it possible to make forecasts based on the estimated multifractional Brownian motion. The application to high-frequency foreign exchange markets (GBP, CHF, SEK, USD, CAD, AUD, JPY, CNY and SGD, all against EUR) shows significantly good forecasts. When the Hurst exponent is higher than 0.5, what depicts a long-memory feature, the accuracy is higher.

[1]  Jean‐François Coeurjolly.  Erratum: Identification of multifractional Brownian motion , 2006 .

[2]  J. L. Véhel,et al.  The Generalized Multifractional Brownian Motion , 2000 .

[3]  Murad S. Taqqu,et al.  MULTIFRACTIONAL PROCESSES WITH RANDOM EXPONENT , 2005 .

[4]  D. Guégan,et al.  Probability density of the empirical wavelet coefficients of a noisy chaos , 2014 .

[5]  Jingtao Yao,et al.  A case study on using neural networks to perform technical forecasting of forex , 2000, Neurocomputing.

[6]  John T. Kent,et al.  Estimating the Fractal Dimension of a Locally Self-similar Gaussian Process by using Increments , 1997 .

[7]  Matthieu Garcin.  Empirical Wavelet Coefficients and Denoising of Chaotic Data in the Phase Space , 2016 .

[8]  B. Mandelbrot,et al.  Fractional Brownian Motions, Fractional Noises and Applications , 1968 .

[9]  Sergio Da Silva,et al.  Hurst exponents, power laws, and efficiency in the Brazilian foreign exchange market , 2007 .

[10]  P. Manimaran,et al.  Multifractal detrended cross-correlation analysis on gold, crude oil and foreign exchange rate time series , 2014 .

[11]  S. Mitra.  Is Hurst Exponent Value Useful in Forecasting Financial Time Series , 2012 .

[12]  D. Grech,et al.  The local Hurst exponent of the financial time series in the vicinity of crashes on the Polish stock exchange market , 2008 .

[13]  Zhanjiang Zhi,et al.  Primal-dual method to smoothing TV-based model for image denoising , 2016 .

[14]  Z. M. Bi,et al.  Time Series Forecasting using Wavelets with Predictor-Corrector Boundary Treatment , 2001 .

[15]  Ke Chen,et al.  Introduction to variational image-processing models and applications , 2013, Int. J. Comput. Math..

[16]  J. Coeurjolly,et al.  HURST EXPONENT ESTIMATION OF LOCALLY SELF-SIMILAR GAUSSIAN PROCESSES USING SAMPLE QUANTILES , 2005, math/0506290.

[17]  J. Álvarez-Ramírez,et al.  Time-varying Hurst exponent for US stock markets , 2008 .

[18]  D. Guégan,et al.  Optimal wavelet shrinkage of a noisy dynamical system with non-linear noise impact , 2015 .

[19]  R. Weron.  Estimating long range dependence: finite sample properties and confidence intervals , 2001, cond-mat/0103510.

[20]  Jean-François Coeurjolly,et al.  Identification of multifractional Brownian motion , 2005 .

[21]  Benoit B. Mandelbrot.  The variation of the prices of cotton, wheat, and railroad stocks, and of some financial rates , 1997 .

[22]  C. Peng,et al.  Mosaic organization of DNA nucleotides. , 1994, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[23]  Robert A. Lordo,et al.  Nonparametric and Semiparametric Models , 2005, Technometrics.

[24]  L. Rudin,et al.  Nonlinear total variation based noise removal algorithms , 1992 .

[25]  R. Peltier,et al.  Multifractional Brownian Motion : Definition and Preliminary Results , 1995 .

[26]  S. Jaffard,et al.  Elliptic gaussian random processes , 1997 .

[27]  S. Liaw,et al.  General and specific statistical properties of foreign exchange markets during a financial crash , 2016 .

[28]  A. Guillin,et al.  Local estimation of the Hurst index of multifractional Brownian motion by Increment Ratio Statistic method , 2010, 1010.4849.

[29]  D. Ghosh,et al.  Multifractal detrended Cross Correlation Analysis of Foreign Exchange and SENSEX fluctuation in Indian perspective , 2016 .

[30]  Semiparametric Bootstrap Approach to Hypothesis Tests and Confidence Intervals for the Hurst Coefficient , 1999 .

[31]  Nonparametric estimation of the local Hurst function of multifractional Gaussian processes , 2010, 1010.2895.

[32]  Howard L. Weinert,et al.  Efficient computation for Whittaker-Henderson smoothing , 2007, Comput. Stat. Data Anal..

[33]  T. Greville.  Moving-Weighted-Average Smoothing Extended to the Extremities of the Data. , 1977 .

[34]  Sergio Bianchi,et al.  Pathwise Identification Of The Memory Function Of Multifractional Brownian Motion With Application To Finance , 2005 .

[35]  Jacques Istas,et al.  Identifying the multifractional function of a Gaussian process , 1998 .

[36]  H. E. Hurst,et al.  Long-Term Storage Capacity of Reservoirs , 1951 .

[37]  J. Geweke,et al.  THE ESTIMATION AND APPLICATION OF LONG MEMORY TIME SERIES MODELS , 1983 .

[38]  T. D. Matteo,et al.  Dynamical generalized Hurst exponent as a tool to monitor unstable periods in financial time series , 2011, 1109.0465.

[39]  Marcel Ausloos,et al.  Statistical physics in foreign exchange currency and stock markets , 2000 .

[40]  B. Mandelbrot.  The Variation of Some Other Speculative Prices , 1967 .

[41]  D. Guégan,et al.  Extreme values of random or chaotic discretization steps , 2012 .

[42]  P. Bertrand,et al.  Modelling NASDAQ Series by Sparse Multifractional Brownian Motion , 2012 .

[43]  Gabriel Lang,et al.  Quadratic variations and estimation of the local Hölder index of a gaussian process , 1997 .

[44]  R. Mantegna,et al.  Scaling behaviour in the dynamics of an economic index , 1995, Nature.

[45]  Pointwise Regularity Exponents and Market Cross-Correlations , 2010 .