How to improve accuracy for DFA technique

TL;DR

This paper investigates how various parameters in Detrended Fluctuation Analysis (DFA) affect the accuracy of estimating the Hurst exponent, using Monte Carlo simulations to evaluate confidence intervals.

Contribution

It provides a detailed analysis of how DFA parameters influence the accuracy of H estimation, offering guidance for parameter selection.

Findings

Parameters like time length L and number of divisors d significantly impact accuracy.

Monte Carlo simulations effectively evaluate confidence intervals for DFA.

Optimal parameter choices improve the reliability of Hurst exponent estimation.

Abstract

This paper extends the existing literature on empirical estimation of the confidence intervals associated to the Detrended Fluctuation Analysis (DFA). We used Montecarlo simulation to evaluate the confidence intervals. Varying the parameters in DFA technique, we point out the relationship between those and the standard deviation of H. The parameters considered are the finite time length L, the number of divisors d used and the values of those. We found that all these parameters play a crucial role, determining the accuracy of the estimation of H.