Multifractal Diffusion Entropy Analysis: Optimal Bin Width of Probability Histograms

Abstract

In the framework of Multifractal Diffusion Entropy Analysis we propose a method for choosing an optimal bin-width in histograms generated from underlying probability distributions of interest. The method presented uses techniques of Rényi's entropy and the mean squared error analysis to discuss the conditions under which the error in the multifractal spectrum estimation is minimal. We illustrate the utility of our approach by focusing on a scaling behavior of financial time series. In particular, we analyze the S&P500 stock index as sampled at a daily rate in the time period 1950-2013. In order to demonstrate a strength of the method proposed we compare the multifractal $\delta$-spectrum for various bin-widths and show the robustness of the method, especially for large values of $q$. For such values, other methods in use, e.g., those based on moment estimation, tend to fail for heavy-tailed data or data with long correlations. Connection between the $\delta$-spectrum and Rényi's $q$ parameter is also discussed and elucidated on a simple example of multiscale time series.

25 pages, 7 figures, LaTeX with the Elsevier article class, substantially revised version

References in corpus (1)

• The world according to Renyi: Thermodynamics of multifractal systems

Cited by in corpus (12)

• Multifractal analysis of financial markets
• On the Role of Information Theoretic Uncertainty Relations in Quantum Theory
• Crucial events, randomness and multi-fractality in heartbeats
• On q-non-extensive statistics with non-Tsallisian entropy
• Rényi and Tsallis entropies: three analytic examples
• On Statistical Properties of Jizba-Arimitsu Hybrid Entropy
• Rescaling the nonadditivity parameter in Tsallis thermostatistics
• Thermodynamics of exponential Kolmogorov-Nagumo averages
• Point Information Gain and Multidimensional Data Analysis
• Point Divergence Gain and Multidimensional Data Sequences Analysis
• Techniques for multifractal spectrum estimation in financial time series
• Causal inference in time series in terms of Rényi transfer entropy