Miwakeichi, F.; Galka, A. Comparison of Bootstrap Methods for Estimating Causality in Linear Dynamic Systems: A Review. Entropy2023, 25, 1070.
Miwakeichi, F.; Galka, A. Comparison of Bootstrap Methods for Estimating Causality in Linear Dynamic Systems: A Review. Entropy 2023, 25, 1070.
Miwakeichi, F.; Galka, A. Comparison of Bootstrap Methods for Estimating Causality in Linear Dynamic Systems: A Review. Entropy2023, 25, 1070.
Miwakeichi, F.; Galka, A. Comparison of Bootstrap Methods for Estimating Causality in Linear Dynamic Systems: A Review. Entropy 2023, 25, 1070.
Abstract
In this study, we meticulously compared the practical performance of four bootstrap methods for assessing the significance of causal analysis in time series data, recognizing that their evaluation has not been sufficiently conducted in the field. The methods investigated were uncorrelated Phase Randomization Bootstrap (uPRB), which generates surrogate data with no cross-correlation between variables by randomizing the phase in the frequency domain; Time Shift Bootstrap (TSB), which generates surrogate data by randomizing the phase in the time domain; Stationary Bootstrap (SB), which calculates standard errors and constructs confidence regions for weakly dependent stationary observations; and AR-sieve bootstrap (ARSB), a resampling method based on autoregressive (AR) models that approximates the underlying data-generating process. Our study found that the AR-sieve bootstrap (ARSB) method outperformed the others in detecting both self-excitation and causality among variables. In contrast, the uncorrelated phase-randomized bootstrap (uPRB) and Stationary Bootstrap (SB) methods demonstrated limitations in specific scenarios. This detailed comparison highlights the need for selecting suitable bootstrap methods to ensure accurate results, ultimately guiding researchers in their choice of method for real data analysis.
Keywords
Causal analysis; Granger Causality; Bootstrap methods; Multivariate time series; Impulse response function
Subject
Computer Science and Mathematics, Signal Processing
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
