{"paper_id":"2401.00507","title":"Optimization of portfolios with cryptocurrencies: Markowitz and GARCH-Copula model approach","authors":["Vahidin Jeleskovic","Claudio Latini","Zahid I. Younas","Mamdouh A. S. Al-Faryan"],"paper_date":"2023-12-31","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2401.00507v1","page_url":"https://thequant.space/flowcharts/optimization-of-portfolios-with-cryptocurrencies-markowitz/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes using Markowitz optimization together with GARCH‑Copula and GARCH‑Vine Copula risk models to build optimal portfolios that include cryptocurrencies and traditional assets.","quote":"We employ the classic variance-covariance approach, utilizing the GARCH-Copula and GARCH-Vine Copula methods to calculate the risk structure. The optimal asset weights within the optimized portfolios are determined through the Markowitz optimization problem.","summary":"The growing interest in cryptocurrencies has drawn the attention of the financial world to this innovative medium of exchange. This study aims to explore the impact of cryptocurrencies on portfolio performance. We conduct our analysis retrospectively, assessing the performance achieved within a…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Financial returns are not assumed to be Gaussian; the authors note that assuming normality underestimates risk.","type":"statistical","stated":"explicit","quote":"Assuming that financial returns are normally distributed lead to their underestimation and inaccurate quantification of risk (Pinudom et al., 2018)."},{"assumption":"The Location‑Scale model is valid for modeling returns, with conditional mean and variance evolving over time.","type":"model","stated":"explicit","quote":"A necessary but not sufficient condition is to accept the assumption that the studies underlying the Location‑Scale model are valid, considering that returns are the variable under the microscope for this analysis."},{"assumption":"Variance of returns changes over time and can be captured by GARCH‑type processes.","type":"model","stated":"explicit","quote":"Empirical evidence (beginning with Engle (1982)) shows that it is possible to consider the variance variant over time."},{"assumption":"Data prior to COVID‑19 and the Ukraine conflict represent a “normal” economic environment for the analysis.","type":"data","stated":"explicit","quote":"Our rationale is rooted in the belief that using data from periods prior to these crises provides a more faithful representation of the ordinary and stable economic environment."},{"assumption":"Missing price data of assets with >10% gaps are excluded and remaining gaps are linearly interpolated.","type":"execution","stated":"explicit","quote":"The missing data were treated with the technique of linear interpolation (Noor et al, 2014). The assets that had more than 10% missing data in their series of historical returns are not considered."}]},"datasets":{"status":"available","items":[{"name":"Cryptocurrency trade volume data (Bitcoin, Ethereum, XRP)","provider":"coinmetrics.io","kind":"crypto-onchain","access":"public","period":"31 Dec 2018 (snapshot)","universe":"Top three cryptocurrencies by market cap","quote":"The three cryptocurrencies with the highest market capitalization (Bitcoin, Etherium and XRP) on 31st December 2018 reached a trade volume equal to 579.582.414 units for an amount of Euro equal to 113.336.458.098,33. Data provided by https://coinmetrics.io."},{"name":"Traditional asset price series","provider":"investing.com","kind":"market-prices","access":"public","period":"1 Jan 2017 to 31 Dec 2018","universe":"Equities, bonds, commodities listed in Table 1","quote":"The data referred to the “Traditional portfolio” are provided by investing.com."},{"name":"Cryptocurrency price series for crypto‑only portfolio","provider":"cryptomarket.com","kind":"market-prices","access":"public","period":"1 Jan 2017 to 31 Dec 2018","universe":"Various cryptocurrencies (e.g., NEO, Decred, Waves, Verge, Unobtanium, Groestlcoin)","quote":"The data are provided by cryptomarket.com."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The portfolio that combines traditional assets with the best‑performing cryptocurrency achieves a higher adjusted Sharpe ratio and more stable performance than either the traditional‑only or crypto‑only portfolios.","kind":"performance","quote":"Our analysis predominantly reveals that the portfolio comprising both cryptocurrency and traditional assets exhibits a higher Sharpe ratio from a retrospective viewpoint and demonstrates more stable performances from a prospective perspective."},{"finding":"The GARCH‑t Copula Markowitz optimization yields the best performance for the Traditional + Crypto portfolio.","kind":"performance","quote":"the best performance is achieved by the “Traditional Crypto portfolio” with the “Markowitz optimization” approach and for “Traditional Crypto portfolio” through the “GARCH-t Copula Markowitz optimization”"},{"finding":"Cryptocurrencies are not excessively riskier than traditional assets, as indicated by comparable or lower risk measures (e.g., kurtosis, standard deviation).","kind":"statistical","quote":"the results in table 1 show that resulting in an averagely higher Adjusted Sharpe Ratio, cryptocurrencies, are not excessively riskier as compare to traditional assets."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":26,"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":"2026-11-14T18:17:11Z"},"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/"}