{"paper_id":"2305.06961","title":"Copula-Based Trading of Cointegrated Cryptocurrency Pairs","authors":["Masood Tadi","Jiří Witzany"],"paper_date":"2023-05-11","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2305.06961v2","page_url":"https://thequant.space/flowcharts/copula-based-trading-of-cointegrated-cryptocurrency-pairs/","scores":{"empirical_rigor":8,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a novel pairs trading strategy that uses copulas to trade cointegrated cryptocurrency pairs.","quote":"This research introduces a novel pairs trading strategy based on copulas for cointegrated pairs of cryptocurrencies.","summary":"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…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"All trades are executed using market orders, incurring taker fees.","type":"execution","stated":"explicit","quote":"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)."},{"assumption":"Maker fee is 0.02% and taker fee is 0.04% on Binance USDT‑Margined Futures.","type":"execution","stated":"explicit","quote":"For instance, on Binance, the maker fee of USDT-Margined Futures is set at 0.02%, while the taker fee is 0.04%."},{"assumption":"If fewer than two stationary spread processes are identified, the strategy abstains from trading.","type":"model","stated":"explicit","quote":"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."},{"assumption":"Log‑returns are replaced with stationary spread processes to ensure mean‑reversion.","type":"statistical","stated":"explicit","quote":"a new methodology is proposed in this study to replace log-returns with stationary spread processes in order to overcome this issue."},{"assumption":"Cointegration is tested using both the linear Engle‑Granger two‑step method and the KSS non‑linear test.","type":"model","stated":"explicit","quote":"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."},{"assumption":"Initial capital of 200,000 USDT is allocated equally to each side of the pair.","type":"execution","stated":"explicit","quote":"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."}]},"datasets":{"status":"available","items":[{"name":"Historical hourly closed prices for twenty cryptocurrency coins","provider":"Binance API","kind":"market-prices","access":"public","period":"01/01/2021 to 19/01/2023","universe":"twenty cryptocurrency coins","quote":"Using the Binance API, we collected historical hourly closed prices for twenty cryptocurrency","known_id":"binance"}],"known":[{"id":"binance","name":"Binance","provider":"Binance","kind":"market-prices","access":"public","quote":"Using the Binance API, we collected historical hourly closed prices for twenty cryptocurrency coins from 01/01/2021 to 19/01/2023.","mentions":10,"used_in":2}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"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.","quote":"For instance, on Binance, the maker fee of USDT-Margined Futures is set at 0.02%, while the taker fee is 0.04%.","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"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.","card_quote":"For instance, on Binance, the maker fee of USDT-Margined Futures is set at 0.02%, while the taker fee is 0.04%.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"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.","kind":"performance","quote":"... Both tests of the pairs trading strategy outperform the Bitcoin buy-and-hold strategy and the portfolio buy-and-hold strategy."},{"finding":"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.","kind":"performance","quote":"The Bitcoin buy-and-hold strategy shows a negative annualized return of −17.0%"},{"finding":"Higher entry thresholds increase volatility and transaction costs but do not improve returns, indicating the optimal threshold is 0.10.","kind":"performance","quote":"When the entry threshold α1 increases, it results in an increase in the number of trading signals. This, in turn, leads to higher volatility"},{"finding":"The copula‑based pairs trading strategy outperforms the portfolio buy‑and‑hold by over six times when α1=0.10.","kind":"performance","quote":"Our pairs trading strategy outperforms the portfolio buy-and-hold by over six times when α1 = 0.10."}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":21,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:27:04Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}