Volatility, irregularity, and predictable degree of accumulative return series

Area of Science:

• Quantitative Finance
• Financial Econometrics
• Complexity Science

Background:

• Financial time series exhibit complex dependencies, challenging the random-walk hypothesis.
• Understanding the link between serial structure and predictability in financial markets is crucial.

Purpose of the Study:

• To propose a framework for magnifying correlations and regularities in financial time series.
• To effectively distinguish real-world financial data from random processes.
• To investigate the relationship between accumulative return series length and market predictability.

Main Methods:

• Construction of accumulative return series to enhance underlying patterns.
• Analysis of volatility, Hurst exponent, and approximate entropy changes.
• Quantification of predictability based on serial structure analysis.

Main Results:

• The proposed framework effectively differentiates financial time series from random walks.
• Predictability of financial time series increases with the length of the accumulative return series.
• Approximate entropy, when adjusted for volatility, serves as a robust indicator of predictability.

Conclusions:

• Financial time series possess a degree of predictability.
• The accumulative return series method offers a valuable tool for analyzing financial data.
• Approximate entropy is a significant metric for assessing financial market predictability.