Paper: arXiv 2411.14058

Authors: Tatsuru Kikuchi

Abstract

In this study, we perform some analysis for the probability distributions in the space of frequency and time variables. However, in the domain of high frequencies, it behaves in such a way as the highly non-linear dynamics. The wavelet analysis is a powerful tool to perform such analysis in order to search for the characteristics of frequency variations over time for the prices of major cryptocurrencies. In fact, the wavelet analysis is found to be quite useful as it examine the validity of the efficient market hypothesis in the weak form, especially for the presence of the cyclical persistence at different frequencies. If we could find some cyclical persistence at different frequencies, that means that there exist some intrinsic causal relationship for some given investment horizons defined by some chosen sampling scales. This is one of the characteristic results of the wavelet analysis in the time-frequency domains.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced mathematical concepts like symplectic geometry and detailed wavelet transform equations (e.g., continuous wavelet transform with Morlet wavelets) on a real dataset (cryptocurrencies and macro assets) with time-series analysis, indicating a mix of theoretical depth and data-driven methodology.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Test Efficient Market Hypothesis<br>for Cryptocurrencies using Wavelet Analysis"] --> B["Methodology<br>Wavelet Analysis on Time-Frequency Domains"]
  B --> C["Data Input<br>High-Frequency Cryptocurrency Price Data"]
  C --> D["Computational Process<br>Apply Wavelet Transform to detect<br>Non-Linear Dynamics and Cyclical Persistence"]
  D --> E["Key Findings<br>Presence of Cyclical Persistence<br>at specific frequencies<br>Weak Form EMH Validation"]

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

The paper proposes using wavelet analysis to examine frequency variations over time for major cryptocurrency prices and to test the weak-form efficient market hypothesis.

Source sentence
The wavelet analysis is a powerful tool to perform such analysis in order to search for the characteristics of frequency variations over time for the prices of major cryptocurrencies. In fact, the wavelet analysis is found to be quite useful as it examine the validity of the efficient market hypothesis in the weak form, especially for the presence of the cyclical persistence at different frequencies.

Assumes 2

  • High‑frequency noisy components are removed by applying high‑frequency filters. model
    Source sentence
    In other words, it is necessary to remove noisy, high-frequency behaviors by introducing high-frequency filters.
  • The daily closing price data obtained via the CryptoCompare API accurately represent market prices. data
    Source sentence
    We extract the historical daily exchange prices of those cryptocurrencies through the API (Application Programming Interface) provided by the CryproCompare, who is one of the leading global cryptocurrency market data providers. Taking such a limitation into account, the study examines the daily closing prices for three major cryptocurrencies (BTC, ETH and 4 XRP) from July 10, 2017 to December 31, 2022 through the

Data

  • Daily cryptocurrency prices (BTC, ETH, XRP) · CryptoCompare market-prices public
    2017-07-10 to 2022-12-31 · Bitcoin, Ethereum, Ripple
    Source sentence
    2 Data and Basic Methodology As similar to our previous studies, we take three major cryptocurrencies, that is, Bitcoin (BTC), Ethereum (ETH) and Ripple (XRP). We extract the historical daily exchange prices of those cryptocurrencies through the API (Application Programming Interface) provided by the CryproCompare, who is one of the leading global cryptocurrency market data providers.
  • S&P 500 index · FRED macro public
    2017-07-10 to 2022-12-31
    Source sentence
    Moreover, we use the historical daily exchange data for S&P500 index and GOLD price, as well as the historical daily exchange data for the JPY/USD (Japanese Yen and US Dollar) and the USD/EUR (US Dollar and Euro) currency pairs with the period of time from July 10, 2017 to December 31, 2022. The data for S&P500 index is extracted from the FRED (Federal Reserve Economic Data) and the data for GOLD/USD, JPY/USD and
  • Gold price (USD) · investing.com macro public
    2017-07-10 to 2022-12-31
    Source sentence
    Moreover, we use the historical daily exchange data for S&P500 index and GOLD price, as well as the historical daily exchange data for the JPY/USD (Japanese Yen and US Dollar) and the USD/EUR (US Dollar and Euro) currency pairs with the period of time from July 10, 2017 to December 31, 2022. The data for S&P500 index is extracted from the FRED (Federal Reserve Economic Data) and the data for GOLD/USD, JPY/USD and
  • JPY/USD exchange rate · investing.com macro public
    2017-07-10 to 2022-12-31
    Source sentence
    Moreover, we use the historical daily exchange data for S&P500 index and GOLD price, as well as the historical daily exchange data for the JPY/USD (Japanese Yen and US Dollar) and the USD/EUR (US Dollar and Euro) currency pairs with the period of time from July 10, 2017 to December 31, 2022. The data for S&P500 index is extracted from the FRED (Federal Reserve Economic Data) and the data for GOLD/USD, JPY/USD and
  • USD/EUR exchange rate · investing.com macro public
    2017-07-10 to 2022-12-31
    Source sentence
    Moreover, we use the historical daily exchange data for S&P500 index and GOLD price, as well as the historical daily exchange data for the JPY/USD (Japanese Yen and US Dollar) and the USD/EUR (US Dollar and Euro) currency pairs with the period of time from July 10, 2017 to December 31, 2022. The data for S&P500 index is extracted from the FRED (Federal Reserve Economic Data) and the data for GOLD/USD, JPY/USD and

Recognised sources: FRED

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • Cyclical persistence at different frequencies was found, especially for the daily exchange data of XRP. statistical
    Source sentence
    We found some cyclical persistence at different frequencies, especially for the daily exchange data of XRP cryptocurrency.
  • The presence of cyclical persistence implies an intrinsic causal relationship for given investment horizons. theoretical
    Source sentence
    It means that there exists some intrinsic causal relationship for some given investment horizons defined by some chosen sampling scales.
  • In the low‑frequency region, price dynamics are stable over time for all examined cryptocurrencies. statistical
    Source sentence
    15 The result of our wavelet analysis shows that if we focus on the low-frequency region, it is stable enough as the time changes for any cryptocurrencies we are interested in.
  • In the high‑frequency region, distinct “hot spots” appear, indicating peaks in the standard time‑series analysis. statistical
    Source sentence
    However, if we take the high-frequency region, that is in the bottom line of each figures, we can see the ’hot spots’ as represented by the red color. Those hot spots corresponds to the peaks in the standard, one-dimensional time series analysis.
  • During the COVID‑19 peak in 2022, all cryptocurrencies exhibit hot spots in the wavelet analysis. statistical
    Source sentence
    One of the most intriguing observation is that during the period of Covid-19 peek in 2022, all the cryptocurrencies show the hot spots in the wavelet analysis.

Checked by others

Supported by

  • Bitcoin: A life in crises (2023)
    On the question: What are the locations of price peaks in Bitcoin high‑frequency data?
    Compare the findings

    This paper: In the high‑frequency region, distinct “hot spots” appear, indicating peaks in the standard time‑series analysis.

    That paper: Price peaks consist of collections of discontinuities, visible as green vertical lines or red areas in the periodograms.

    Both detect peaks: A as discontinuities (green lines/red areas) in periodograms, B as hot‑spots in the high‑frequency region.

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.