Paper: arXiv 2305.06961
Authors: Masood Tadi, Jiří Witzany
Abstract
This research introduces a novel pairs trading strategy based on copulas for cointegrated pairs of cryptocurrencies. To identify the most suitable pairs, the study employs linear and non-linear cointegration tests along with a correlation coefficient measure and fits different copula families to generate trading signals formulated from a reference asset for analyzing the mispricing index. The strategy’s performance is then evaluated by conducting back-testing for various triggers of opening positions, assessing its returns and risks. The findings indicate that the proposed method outperforms buy-and-hold trading strategies in terms of both profitability and risk-adjusted returns.
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 mathematical concepts like copula theory, non-linear cointegration tests, and conditional probability derivatives, indicating high mathematical complexity. It also demonstrates high empirical rigor by conducting back-tests on historical cryptocurrency data, assessing returns and risks, and comparing performance against buy-and-hold strategies.
Research Flowchart
flowchart TD A["Research Goal<br>Develop a copula-based pairs trading strategy<br>for cointegrated cryptocurrency pairs"] --> B["Data Collection & Preparation"] B --> C["Pair Selection"] C --> D["Model Fitting<br>Copula Family Selection"] D --> E["Signal Generation<br>Mispricing Index & Trading Triggers"] E --> F["Back-testing & Evaluation<br>Performance & Risk Metrics"] F --> G["Key Outcomes<br>Strategy outperforms Buy-and-Hold<br>in profitability & risk-adjusted returns"] B["Data Preparation"] --> B1["Historical Crypto Price Data"] C["Pair Selection"] --> C1["Linear/Non-linear<br>Cointegration Tests"] C --> C2["Correlation Coefficient"]
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 a novel pairs trading strategy that uses copulas to trade cointegrated cryptocurrency pairs.
Source sentence
This research introduces a novel pairs trading strategy based on copulas for cointegrated pairs of cryptocurrencies.
Assumes 6
- All trades are executed using market orders, incurring taker fees. execution
Source sentence
It is also assumed that all trades are executed using market orders, which typically incur higher fees (known as taker fees) compared to the lower fees charged for limit orders (known as maker fees).
- Maker fee is 0.02% and taker fee is 0.04% on Binance USDT‑Margined Futures. execution
Source sentence
For instance, on Binance, the maker fee of USDT-Margined Futures is set at 0.02%, while the taker fee is 0.04%.
- If fewer than two stationary spread processes are identified, the strategy abstains from trading. model
Source sentence
If we fail to identify a minimum of two stationary spread processes within a specified formation period, we abstain from trading during its corresponding trading period.
- Log‑returns are replaced with stationary spread processes to ensure mean‑reversion. statistical
Source sentence
a new methodology is proposed in this study to replace log-returns with stationary spread processes in order to overcome this issue.
Show 2 more
- Cointegration is tested using both the linear Engle‑Granger two‑step method and the KSS non‑linear test. model
Source sentence
We used the linear Engle‑Granger (EG) two‑step method and to identify cointegrated pairs and calculated Kendall’s Tau. Then, we calibrated the method’s parameters using the copula approach and generated trading signals for the strategy.
- Initial capital of 200,000 USDT is allocated equally to each side of the pair. execution
Source sentence
we initially invested 200,000 USDT, with the coins’ weights set to ensure that each side has a maximum initial capital of around 200,000 USDT.
Data
- Historical hourly closed prices for twenty cryptocurrency coins market-prices public
Source sentence
Using the Binance API, we collected historical hourly closed prices for twenty cryptocurrency
Costs
Modeled
Maker fee 0.02%, taker fee 0.04%; transaction costs over gross P&L range from 11.7% to 47.7% depending on entry threshold.
Source sentence
For instance, on Binance, the maker fee of USDT-Margined Futures is set at 0.02%, while the taker fee is 0.04%.
Finds 4
- Pairs trading with the EG test at entry threshold α1=0.10 yields a total return of 76.2%, annualized Sharpe ratio of 0.97 and outperforms both Bitcoin and portfolio buy‑and‑hold strategies. performance
Source sentence
... Both tests of the pairs trading strategy outperform the Bitcoin buy-and-hold strategy and the portfolio buy-and-hold strategy.
- Bitcoin buy‑and‑hold shows a negative annualized return of –17.0% and a Sharpe ratio of –0.23, while the portfolio buy‑and‑hold returns 14.4% with a Sharpe ratio of 0.15. performance
Source sentence
The Bitcoin buy-and-hold strategy shows a negative annualized return of −17.0%
- Higher entry thresholds increase volatility and transaction costs but do not improve returns, indicating the optimal threshold is 0.10. performance
Source sentence
When the entry threshold α1 increases, it results in an increase in the number of trading signals. This, in turn, leads to higher volatility
- The copula‑based pairs trading strategy outperforms the portfolio buy‑and‑hold by over six times when α1=0.10. performance
Source sentence
Our pairs trading strategy outperforms the portfolio buy-and-hold by over six times when α1 = 0.10.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.
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.