Paper: arXiv 2511.03314
Authors: Tetsuya Takaishi
Abstract
The finite sample effect on the Hurst exponent (HE) of realized volatility time series is examined using Bitcoin data. This study finds that the HE decreases as the sampling period \(Δ\) increases and a simple finite sample ansatz closely fits the HE data. We obtain values of the HE as \(Δ\rightarrow 0\), which are smaller than 1/2, indicating rough volatility. The relative error is found to be $1%$ for the widely used five-minute realized volatility. Performing a multifractal analysis, we find the multifractality in the realized volatility time series, smaller than that of the price-return time series.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced multifractal analysis with specialized estimators (MFDFA) and mathematical derivations for finite sample effects, but is heavily grounded in empirical Bitcoin data with detailed backtesting procedures for realized volatility.
Research Flowchart
flowchart TD A["Research Goal: Examine finite sample effect on Hurst Exponent of Bitcoin realized volatility"] B["Data Input: Bitcoin High-Frequency Price Data"] C["Methodology: Compute Realized Volatility at various sampling intervals Δ"] D["Computational Process 1: Calculate Hurst Exponent for each Δ"] E["Computational Process 2: Perform Multifractal Analysis"] F["Outcome 1: HE decreases as Δ increases; HE→0 is < 0.5 (Rough Volatility)"] G["Outcome 2: Multifractality lower than price returns; 5-min RV relative error 1%"] A --> B B --> C C --> D C --> E D --> F E --> G
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
We provide a simple two‑parameter ansatz that obtains the Hurst exponent at Δ→0 and use it to estimate HE≈0.12‑0.14, indicating roughness of the realized volatility time series, and we perform a multifractal analysis of the five‑minute realized volatility.
Source sentence
10 We provide a simple two-parameter ansatz that obtains the HE at ∆→0 and use ansatz to obtain the HE ∼0.12 −0.14, indicating the roughness of the RV time series.
Assumes 2
- The observed daily return is modeled as rt = σt εt with εt drawn from a standard normal distribution. model
Source sentence
Let us assume that the observed daily return rt is described by rt = σtǫt, where ǫt ∼N(0, 1) and σt is the standard deviation.
- Realized volatility is constructed by summing squared intraday returns sampled at a fixed Δ‑minute frequency (RVt,Δ = Σi ri,Δ²). data
Source sentence
We construct the daily RV on day t with ∆-minute period by RVt,∆= n X i r2 t,i,∆, (1) where rt,i,∆, i = 1, 2, . . . , n are intraday returns and n = 1440/∆is the num-ber of samples in one day.
Data
- Bitcoin tick data crypto-onchain public
Source sentence
Data and Methodology Our data consists of Bitcoin tick data traded on Bitstamp exchange from January 2, 2014, to June 01, 2023.
Costs
Not applicable
Finds 5
- The Hurst exponent decreases as the sampling period Δ increases. statistical
Source sentence
4. Conclusion We examine the finite sample effect on the HE using Bitcoin data and find that the HE decreases as ∆increases.
- The relative error of the Hurst exponent for the widely used five‑minute realized volatility is around 1%. statistical
Source sentence
The relative errors at five-minute ∆are found to be around 1%, suggesting that the finite size effect on the HE for the widely used five-minute RV is reasonably small .
- Multifractality is present in the five‑minute realized volatility series (h(q) varies with q). statistical
Source sentence
It indicates that h(q) as a function of q is not constant, suggesting the multifractality of the time series.
- The strength of multifractality in the realized volatility is smaller than that of the price‑return series and varies over time. statistical
Source sentence
Analyzing the five-minute RV time series, we find that its multifractality is smaller than that of price-return and the strength of the multifractality varies over time.
- The simple two‑parameter ansatz yields an estimated Hurst exponent of approximately 0.12‑0.14, indicating rough volatility of the realized volatility time series. theoretical
Source sentence
10 We provide a simple two-parameter ansatz that obtains the HE at ∆→0 and use ansatz to obtain the HE ∼0.12 −0.14, indicating the roughness of the RV time series.
Checked by others
Supported by
- Rough volatility: evidence from range volatility estimators
Compare the findings
This paper: The simple two‑parameter ansatz yields an estimated Hurst exponent of approximately 0.12‑0.14, indicating rough volatility of the realized volatility time series.
That paper: The Hurst exponent for the range‑based proxies can be as low as 0.014, supporting the rough volatility hypothesis.
- Roughness Analysis of Realized Volatility and VIX through Randomized Kolmogorov-Smirnov Distribution
Compare the findings
This paper: The simple two‑parameter ansatz yields an estimated Hurst exponent of approximately 0.12‑0.14, indicating rough volatility of the realized volatility time series.
That paper: Both volatility measures exhibit Hurst exponents significantly below 1/2, confirming pronounced roughness in volatility dynamics.
- Probabilistic models and statistics for electronic financial markets in the digital age
Compare the findings
This paper: The simple two‑parameter ansatz yields an estimated Hurst exponent of approximately 0.12‑0.14, indicating rough volatility of the realized volatility time series.
That paper: Empirical analysis of the S&P 500 volatility series yields an estimated Hurst exponent ˆH≈0.16, supporting the rough volatility paradigm.
Qualified by
- Multifractality in Bitcoin Realised Volatility: Implications for Rough Volatility Modelling
Compare the findings
This paper: The strength of multifractality in the realized volatility is smaller than that of the price‑return series and varies over time.
That paper: Multifractal Detrended Fluctuation Analysis, log‑log moment scaling, and wavelet leaders consistently indicate a robust multifractal temporal structure in Bitcoin realised volatility.
1 paper in the archive
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.