Optimising cryptocurrency portfolios through stable clustering of price correlation networks

Luis Enrique Correa Rocha; Ruixue Jing; Ryota Kobayashi

arxiv: 2505.24831 · v2 · submitted 2025-05-30 · ⚛️ physics.pop-ph · physics.soc-ph· q-fin.PM

Optimising cryptocurrency portfolios through stable clustering of price correlation networks

Ruixue Jing , Ryota Kobayashi , Luis Enrique Correa Rocha This is my paper

Pith reviewed 2026-05-19 12:42 UTC · model grok-4.3

classification ⚛️ physics.pop-ph physics.soc-phq-fin.PM

keywords cryptocurrencyportfolio optimizationcorrelation networkscommunity detectionLouvain algorithmconsensus clusteringARIMA forecastingtail risk

0 comments

The pith

Stable clusters from cryptocurrency price correlation networks produce portfolios with consistent positive returns and tighter tail-risk control up to 14-day horizons.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that grouping cryptocurrencies into persistent clusters based on their price correlations allows construction of diversified portfolios that deliver steady gains while limiting downside exposure. By detecting communities in correlation networks and refining them through consensus methods plus forward price forecasts, the approach identifies groups whose co-movements remain reliable over short periods. A sympathetic reader would care because cryptocurrency prices swing sharply, and this method offers a repeatable way to select holdings that exploit real interdependencies rather than treating each coin in isolation. The results indicate that these clusters support profitable strategies without requiring constant adjustment, particularly for holding periods of one to two weeks.

Core claim

Using five years of daily closing prices, the authors apply the Louvain community detection algorithm to cryptocurrency correlation networks and then apply consensus clustering to isolate temporally persistent groups of coins. They integrate ARIMA-based price forecasts to make cluster assignment forward-looking, then construct and evaluate portfolios across multiple strategies and holding horizons. The resulting predictive consensus-clustering portfolios maintain consistently positive and stable performance up to a 14-day horizon, show favorable gain-loss asymmetry, and achieve tighter tail-risk control than comparison approaches.

What carries the argument

The central mechanism is Louvain community detection combined with consensus clustering on rolling price-correlation networks, augmented by ARIMA forecasts to form stable, predictive clusters for portfolio weighting.

If this is right

Portfolios built from these clusters deliver positive returns across holding periods from one to fourteen days.
The approach produces better gain-loss asymmetry than portfolios formed without clustering.
Tail-risk measures such as maximum drawdown and value-at-risk improve relative to non-clustered benchmarks.
Stable interdependencies in the cryptocurrency market can be systematically exploited for short-term risk-aware allocation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same clustering logic might extend to intraday or weekly data to test whether cluster stability holds at finer time scales.
If clusters prove durable, the method could reduce transaction costs in live trading by lowering the frequency of portfolio rebalancing.
Applying the framework to mixed asset classes could reveal whether cross-market correlation clusters offer similar diversification benefits.

Load-bearing premise

The extracted correlation clusters remain stable over time and their future co-movements can be anticipated accurately enough by ARIMA forecasts to guide profitable portfolio decisions.

What would settle it

Out-of-sample testing on a later period of cryptocurrency prices where the identified clusters lose correlation stability or the portfolios show negative or unstable returns beyond the 14-day mark would refute the central claim.

Figures

Figures reproduced from arXiv: 2505.24831 by Luis Enrique Correa Rocha, Ruixue Jing, Ryota Kobayashi.

**Figure 1.** Figure 1: The workflow of our portfolio construction method, including price log return prediction (step 1), community detection (step 2), stable [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

**Figure 2.** Figure 2: Flowchart illustrating the strategies employed for detecting stable clusters of highly correlated cryptocurrencies. The strategies are [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Figure 3: Time series of log returns for two cryptocurrencies (A) BTC and (B) ETH for each model of predicting horizon from 1 to 14 days, using [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗

**Figure 4.** Figure 4: The top 20 cryptocurrencies with the largest average of the median MSE from the ARIMA method over all [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗

**Figure 5.** Figure 5: Size distribution of the final stable cryptocurrency clusters after applying the (A) Baseline strategy and (B) P(ARIMA) strategy, during [PITH_FULL_IMAGE:figures/full_fig_p015_5.png] view at source ↗

**Figure 6.** Figure 6: Top 20 cryptocurrencies ranked by the proportion of occurrence within the largest cluster identified across all [PITH_FULL_IMAGE:figures/full_fig_p016_6.png] view at source ↗

**Figure 7.** Figure 7: Performance indicators for our strategies (Baseline and P(ARIMA)) compared with two benchmark methods: a Planar Maximally Filtered [PITH_FULL_IMAGE:figures/full_fig_p018_7.png] view at source ↗

read the original abstract

The rapidly evolving cryptocurrency market presents unique challenges for investment due to its inherent volatility and evolving regulatory environment. Collective price movements can be exploited to construct diversified portfolios with improved risk-return profiles. This paper introduces an integrated framework that combines network analysis, price forecasting, and portfolio theory to identify stable groups of highly correlated cryptocurrencies for profitable portfolio construction. We employ the Louvain community detection algorithm together with consensus clustering to extract temporally persistent correlation clusters, and incorporate ARIMA-based price forecasts to enhance forward-looking cluster formation. Using 5 years of daily closing prices, we evaluate portfolio performance across multiple strategies and holding horizons, assessing both profitability and downside risk with return-based and tail-risk metrics. Our empirical results show that predictive consensus-clustering portfolios maintain consistently positive and stable performance up to a 14-day horizon, exhibit favourable gain-loss asymmetry, and achieve tighter tail-risk control. These findings demonstrate that stable interdependencies in cryptocurrency markets can be leveraged to construct profitable and risk-aware portfolios across short-term holding horizons.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Applies Louvain plus consensus clustering and ARIMA to crypto correlations for short-horizon portfolios, but the out-of-sample stability of those clusters is not clearly demonstrated.

read the letter

This paper takes Louvain community detection and consensus clustering on rolling price correlation networks for cryptocurrencies, adds ARIMA forecasts, and reports that the resulting portfolios show positive returns with decent risk control up to 14-day horizons. The core move is treating the detected clusters as stable groups for diversification and forward-looking allocation. They run this on five years of daily closing prices and compare strategies across holding periods while tracking both returns and tail risk. That setup is straightforward and uses real market data rather than synthetic examples. The addition of consensus clustering to stabilize the groups is a practical choice over single Louvain runs. The evaluation covers multiple horizons and includes downside metrics, which is better than stopping at raw returns. The main limitation is that the headline performance rests on the clusters remaining predictive out of sample, yet the work does not supply direct checks such as adjusted Rand index or variation of information between clusters formed on past windows and realized co-movements ahead. Without those or explicit baseline comparisons to simple correlation-threshold or momentum portfolios, it is hard to separate the contribution of the clustering step from short-term autocorrelation already present in the data. The abstract also omits specific performance numbers, error bars, or data-cleaning rules, which leaves the strength of the empirical claim difficult to gauge. This is the kind of applied paper that might interest people building quantitative strategies for volatile assets like crypto who already know network methods. A reader looking for a replicable recipe could extract the overall workflow, though they would still need to add their own robustness tests. The paper engages honestly with the literature on community detection and forecasting, so it is coherent on its own terms and worth sending to a serious referee who can check the code, data splits, and any unreported stability diagnostics.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces a framework for cryptocurrency portfolio optimization that integrates Louvain community detection and consensus clustering on rolling price correlation networks with ARIMA-based forecasts to identify temporally stable clusters. Using five years of daily closing prices, it evaluates multiple portfolio strategies across holding horizons and claims that the predictive consensus-clustering portfolios deliver consistently positive returns up to 14 days, with favorable gain-loss asymmetry and tighter tail-risk control compared to alternatives.

Significance. If the empirical advantages prove robust under proper validation, the approach could contribute to network-based methods in volatile asset markets by leveraging persistent interdependencies for short-term portfolio construction. The emphasis on consensus clustering to promote stability and the combination with forecasting represent a reasonable extension of existing techniques, though the lack of detailed quantitative benchmarks limits its immediate applicability.

major comments (2)

[Empirical results] Empirical results section: The central claim of 'consistently positive and stable performance' and 'tighter tail-risk control' is not supported by specific quantitative metrics such as average returns, Sharpe ratios, maximum drawdowns, or statistical significance tests against baselines (e.g., equal-weighted or ARIMA-only portfolios). Without these, it is impossible to assess the magnitude or robustness of the reported improvements.
[Methods and results on cluster stability] Cluster formation and stability subsection: No out-of-sample quantitative metric for temporal cluster stability is reported (e.g., adjusted Rand index or variation of information between clusters formed on rolling windows [t-T, t] and realized co-movements on [t+1, t+H]). This is load-bearing for the claim that the clusters are predictive rather than artifacts of short-term autocorrelation, as the headline performance could otherwise be driven by data snooping.

minor comments (2)

[Abstract] The abstract refers to 'favourable gain-loss asymmetry' without defining the exact metric (e.g., upside/downside potential ratio or similar); this should be specified for reproducibility.
[Methods] Notation for the correlation matrix and consensus clustering parameters (e.g., number of Louvain runs or agreement threshold) could be made more explicit in the methods to aid replication.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their insightful comments, which have helped us identify areas where the manuscript can be strengthened. We address each of the major comments below and outline the revisions we will make.

read point-by-point responses

Referee: Empirical results section: The central claim of 'consistently positive and stable performance' and 'tighter tail-risk control' is not supported by specific quantitative metrics such as average returns, Sharpe ratios, maximum drawdowns, or statistical significance tests against baselines (e.g., equal-weighted or ARIMA-only portfolios). Without these, it is impossible to assess the magnitude or robustness of the reported improvements.

Authors: We agree that providing specific quantitative metrics would enhance the clarity and verifiability of our empirical claims. In the revised version, we will add detailed performance tables including average returns, Sharpe ratios, maximum drawdowns, and statistical significance tests (e.g., paired t-tests) against the specified baselines. This will allow readers to better evaluate the magnitude and robustness of the improvements. revision: yes
Referee: Cluster formation and stability subsection: No out-of-sample quantitative metric for temporal cluster stability is reported (e.g., adjusted Rand index or variation of information between clusters formed on rolling windows [t-T, t] and realized co-movements on [t+1, t+H]). This is load-bearing for the claim that the clusters are predictive rather than artifacts of short-term autocorrelation, as the headline performance could otherwise be driven by data snooping.

Authors: We recognize the importance of an out-of-sample stability metric to substantiate the predictive nature of the clusters. We will incorporate quantitative measures such as the adjusted Rand index comparing clusters from the rolling window [t-T, t] to the realized co-movements in [t+1, t+H]. This addition will help address concerns regarding data snooping and confirm the temporal stability of the identified clusters. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical backtesting of clustering-based portfolios

full rationale

The paper presents a standard empirical pipeline: rolling-window correlation matrices, Louvain + consensus clustering, ARIMA forecasts, and out-of-sample portfolio backtesting with return and tail-risk metrics. No equation or result is shown to reduce by construction to a fitted parameter renamed as a prediction, nor does any load-bearing claim rest on a self-citation chain that itself assumes the target result. Performance claims are evaluated against external benchmarks (Sharpe, VaR, etc.) rather than derived tautologically from the clustering procedure itself.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review reveals no explicit free parameters, axioms, or invented entities; the work rests on standard algorithms whose assumptions are not enumerated here.

pith-pipeline@v0.9.0 · 5707 in / 1120 out tokens · 56991 ms · 2026-05-19T12:42:46.341300+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

51 extracted references · 51 canonical work pages

[1]

ElBahrawy, L

A. ElBahrawy, L. Alessandretti, A. Kandler, R. Pastor-Satorras, A. Baronchelli, Evolutionary dynamics of the cryptocurrency market, Royal Society open science 4 (11) (2017) 170623

work page 2017
[2]

F. Y . Legotin, A. A. Kocherbaeva, V . E. Savin, Prospects for crypto-currency and blockchain technologies in financial markets, Revista Espacios 39 (19) (2018)

work page 2018
[3]

Antonakakis, I

N. Antonakakis, I. Chatziantoniou, D. Gabauer, Cryptocurrency market contagion: Market uncertainty, mar- ket complexity, and dynamic portfolios, Journal of International Financial Markets, Institutions and Money 61 (2019) 37–51

work page 2019
[4]

Corbet, B

S. Corbet, B. Lucey, A. Urquhart, L. Yarovaya, Cryptocurrencies as a financial asset: A systematic analysis, International Review of Financial Analysis 62 (2019) 182–199

work page 2019
[5]

H. Arslanian, The emergence of new blockchains and crypto-assets, in: The Book of Crypto: The Complete Guide to Understanding Bitcoin, Cryptocurrencies and Digital Assets, Springer, 2022, pp. 99–119

work page 2022
[6]

S. Arsi, S. Ben Khelifa, Y . Ghabri, H. Mzoughi, Cryptocurrencies: Key risks and challenges, in: Cryptofinance: A New Currency for a New Economy, World Scientific, 2022, pp. 121–145

work page 2022
[7]

Msefula, T

G. Msefula, T. C.-T. Hou, T. Lemesi, Financial and market risks of bitcoin adoption as legal tender: evidence from el salvador, Humanities and Social Sciences Communications 11 (1) (2024) 1–15

work page 2024
[8]

T. Li, D. Shin, B. Wang, Cryptocurrency pump-and-dump schemes, Journal of Financial and Quantitative Anal- ysis (2018) 1–59

work page 2018
[9]

Pintelas, I

E. Pintelas, I. E. Livieris, S. Stavroyiannis, T. Kotsilieris, P. Pintelas, Investigating the problem of cryptocurrency price prediction: a deep learning approach, in: Artificial Intelligence Applications and Innovations: 16th IFIP 23 WG 12.5 International Conference, AIAI 2020, Neos Marmaras, Greece, June 5–7, 2020, Proceedings, Part II 16, Springer, 202...

work page 2020
[10]

Yang, X.-Y

L. Yang, X.-Y . Liu, X. Li, Y . Li, Price prediction of cryptocurrency: an empirical study, in: Smart Blockchain: Second International Conference, SmartBlock 2019, Birmingham, UK, October 11–13, 2019, Proceedings 2, Springer, 2019, pp. 130–139

work page 2019
[11]

A. M. Khedr, I. Arif, M. El-Bannany, S. M. Alhashmi, M. Sreedharan, Cryptocurrency price prediction using traditional statistical and machine-learning techniques: A survey, Intelligent Systems in Accounting, Finance and Management 28 (1) (2021) 3–34

work page 2021
[12]

M. M. Patel, S. Tanwar, R. Gupta, N. Kumar, A deep learning-based cryptocurrency price prediction scheme for financial institutions, Journal of information security and applications 55 (2020) 102583

work page 2020
[13]

Vikram, N

K. Vikram, N. Sivaraman, P. Balamurugan, Crypto currency market price prediction using data science process, International Journal for Research in Applied Science and Engineering Technology 10 (2) (2022) 1451–1454

work page 2022
[14]

Deprez, M

N. Deprez, M. Frömmel, Are simple technical trading rules profitable in bitcoin markets?, International Review of Economics & Finance 93 (2024) 858–874

work page 2024
[15]

W. N. Goetzmann, A. Kumar, Equity portfolio diversification, Review of Finance 12 (3) (2008) 433–463

work page 2008
[16]

Basak, A

S. Basak, A. Pavlova, A. Shapiro, Optimal asset allocation and risk shifting in money management, The Review of Financial Studies 20 (5) (2007) 1583–1621

work page 2007
[17]

Brauneis, R

A. Brauneis, R. Mestel, Cryptocurrency-portfolios in a mean-variance framework, Finance Research Letters 28 (2019) 259–264

work page 2019
[18]

F. J. Fabozzi, H. M. Markowitz, F. Gupta, Portfolio selection, Handbook of Finance 2 (2008)

work page 2008
[19]

Chaves, J

D. Chaves, J. Hsu, F. Li, O. Shakernia, Risk parity portfolio vs. other asset allocation heuristic portfolios, Journal of Investing 20 (1) (2011) 108

work page 2011
[20]

Amirzadeh, A

R. Amirzadeh, A. Nazari, D. Thiruvady, Applying artificial intelligence in cryptocurrency markets: A survey, Algorithms 15 (11) (2022) 428

work page 2022
[21]

Jiang, J

Z. Jiang, J. Liang, Cryptocurrency portfolio management with deep reinforcement learning, in: 2017 Intelligent systems conference (IntelliSys), IEEE, 2017, pp. 905–913. 24

work page 2017
[22]

Karalevicius, N

V . Karalevicius, N. Degrande, J. De Weerdt, Using sentiment analysis to predict interday bitcoin price move- ments, The Journal of Risk Finance 19 (1) (2018) 56–75

work page 2018
[23]

Kim, S.-Y

K. Kim, S.-Y . T. Lee, S. Assar, The dynamics of cryptocurrency market behavior: sentiment analysis using markov chains, Industrial Management & Data Systems 122 (2) (2022) 365–395

work page 2022
[24]

P. Rao, N. Bhasin, P. Goswami, L. Aggarwal, Crypto currency portfolio allocation using machine learning, in: 2021 3rd International Conference on Advances in Computing, Communication Control and Networking (ICAC3N), IEEE, 2021, pp. 1522–1527

work page 2021
[25]

E. W. V . Zuniga, C. M. Ranieri, L. Zhao, J. Ueyama, Y .-t. Zhu, D. Ji, Maximizing portfolio profitability during a cryptocurrency downtrend: A bitcoin blockchain transaction-based approach, Procedia Computer Science 222 (2023) 539–548

work page 2023
[26]

T. Lan, M. Frömmel, Risk factors in cryptocurrency pricing, International Review of Financial Analysis (2025) 104389

work page 2025
[27]

M. Y . Hong, J. W. Yoon, The impact of covid-19 on cryptocurrency markets: A network analysis based on mutual information, Plos ONE 17 (2) (2022) e0259869

work page 2022
[28]

A. Rea, W. Rea, Visualization of a stock market correlation matrix, Physica A: Statistical Mechanics and its Applications 400 (2014) 109–123

work page 2014
[29]

Lyócsa, T

Š. Lyócsa, T. V `yrost, E. Baumöhl, Stock market networks: The dynamic conditional correlation approach, Physica A: Statistical Mechanics and its Applications 391 (16) (2012) 4147–4158

work page 2012
[30]

Onnela, A

J.-P. Onnela, A. Chakraborti, K. Kaski, J. Kertesz, A. Kanto, Dynamics of market correlations: Taxonomy and portfolio analysis, Physical Review E 68 (5) (2003) 056110

work page 2003
[31]

R. Jing, L. E. Rocha, A network-based strategy of price correlations for optimal cryptocurrency portfolios, Finance Research Letters 58 (2023) 104503

work page 2023
[32]

Ioannidis, I

E. Ioannidis, I. Sarikeisoglou, G. Angelidis, Portfolio construction: A network approach, Mathematics 11 (22) (2023) 4670

work page 2023
[33]

Tumminello, T

M. Tumminello, T. Aste, T. Di Matteo, R. N. Mantegna, A tool for filtering information in complex systems, Proceedings of the National Academy of Sciences 102 (30) (2005) 10421–10426. 25

work page 2005
[34]

A. J. McNeil, R. Frey, P. Embrechts, Quantitative risk management: concepts, techniques and tools-revised edition, Princeton university press, 2015

work page 2015
[35]

P. J. Kaufman, Trading systems and methods, V ol. 591, John Wiley & Sons, 2013

work page 2013
[36]

Pardo, The evaluation and optimization of trading strategies, John Wiley & Sons, 2011

R. Pardo, The evaluation and optimization of trading strategies, John Wiley & Sons, 2011

work page 2011
[37]

V . D. Blondel, J.-L. Guillaume, R. Lambiotte, E. Lefebvre, Fast unfolding of communities in large networks, Journal of statistical mechanics: theory and experiment 2008 (10) (2008) P10008

work page 2008
[38]

Lancichinetti, S

A. Lancichinetti, S. Fortunato, Consensus clustering in complex networks, Scientific reports 2 (1) (2012) 336

work page 2012
[39]

Akaike, A new look at the statistical model identification, IEEE transactions on automatic control 19 (6) (1974) 716–723

H. Akaike, A new look at the statistical model identification, IEEE transactions on automatic control 19 (6) (1974) 716–723

work page 1974
[40]

Abadi, P

M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al., Tensorflow: a system for large-scale machine learning, in: 12th USENIX symposium on operating systems design and implementation (OSDI 16), 2016, pp. 265–283

work page 2016
[41]

O. B. Sezer, M. U. Gudelek, A. M. Ozbayoglu, Financial time series forecasting with deep learning: A systematic literature review: 2005–2019, Applied soft computing 90 (2020) 106181

work page 2005
[42]

L. E. Rocha, N. Masuda, P. Holme, Sampling of temporal networks: Methods and biases, Physical Review E 96 (5) (2017) 052302

work page 2017
[43]

R. N. Mantegna, Hierarchical structure in financial markets, The European Physical Journal B 11 (1) (1999) 193–197

work page 1999
[44]

Pantaleo, M

E. Pantaleo, M. Tumminello, F. Lillo, R. N. Mantegna, When do improved covariance matrix estimators enhance portfolio optimization? an empirical comparative study of nine estimators, Quantitative Finance 11 (7) (2011) 1067–1080

work page 2011
[45]

Ledoit, M

O. Ledoit, M. Wolf, A well-conditioned estimator for large-dimensional covariance matrices, Journal of Multi- variate Analysis 88 (2) (2004) 365–411

work page 2004
[46]

R. A. Martin, Pyportfolioopt: portfolio optimization in python, Journal of Open Source Software 6 (61) (2021) 3066

work page 2021
[47]

Dowd, Measuring market risk, John Wiley & Sons, 2007

K. Dowd, Measuring market risk, John Wiley & Sons, 2007. 26

work page 2007
[48]

V . V . Acharya, L. H. Pedersen, T. Philippon, M. Richardson, Measuring systemic risk, The review of financial studies 30 (1) (2017) 2–47

work page 2017
[49]

Keating, W

C. Keating, W. F. Shadwick, An introduction to omega, AIMA newsletter (2002)

work page 2002
[50]

M. I. Rakib, A. Nobi, J. W. Lee, Structure and dynamics of financial networks by feature ranking method, Scientific Reports 11 (1) (2021) 17618

work page 2021
[51]

T. Aste, T. Di Matteo, M. Tumminello, R. N. Mantegna, Correlation filtering in financial time series, in: Noise and Fluctuations in Econophysics and Finance, V ol. 5848, SPIE, 2005, pp. 100–109. 27 Supplementary Information Data source The daily price data of the cryptocurrencies used in our study was obtained from the following websites: www.investing.co...

work page 2005

[1] [1]

ElBahrawy, L

A. ElBahrawy, L. Alessandretti, A. Kandler, R. Pastor-Satorras, A. Baronchelli, Evolutionary dynamics of the cryptocurrency market, Royal Society open science 4 (11) (2017) 170623

work page 2017

[2] [2]

F. Y . Legotin, A. A. Kocherbaeva, V . E. Savin, Prospects for crypto-currency and blockchain technologies in financial markets, Revista Espacios 39 (19) (2018)

work page 2018

[3] [3]

Antonakakis, I

N. Antonakakis, I. Chatziantoniou, D. Gabauer, Cryptocurrency market contagion: Market uncertainty, mar- ket complexity, and dynamic portfolios, Journal of International Financial Markets, Institutions and Money 61 (2019) 37–51

work page 2019

[4] [4]

Corbet, B

S. Corbet, B. Lucey, A. Urquhart, L. Yarovaya, Cryptocurrencies as a financial asset: A systematic analysis, International Review of Financial Analysis 62 (2019) 182–199

work page 2019

[5] [5]

H. Arslanian, The emergence of new blockchains and crypto-assets, in: The Book of Crypto: The Complete Guide to Understanding Bitcoin, Cryptocurrencies and Digital Assets, Springer, 2022, pp. 99–119

work page 2022

[6] [6]

S. Arsi, S. Ben Khelifa, Y . Ghabri, H. Mzoughi, Cryptocurrencies: Key risks and challenges, in: Cryptofinance: A New Currency for a New Economy, World Scientific, 2022, pp. 121–145

work page 2022

[7] [7]

Msefula, T

G. Msefula, T. C.-T. Hou, T. Lemesi, Financial and market risks of bitcoin adoption as legal tender: evidence from el salvador, Humanities and Social Sciences Communications 11 (1) (2024) 1–15

work page 2024

[8] [8]

T. Li, D. Shin, B. Wang, Cryptocurrency pump-and-dump schemes, Journal of Financial and Quantitative Anal- ysis (2018) 1–59

work page 2018

[9] [9]

Pintelas, I

E. Pintelas, I. E. Livieris, S. Stavroyiannis, T. Kotsilieris, P. Pintelas, Investigating the problem of cryptocurrency price prediction: a deep learning approach, in: Artificial Intelligence Applications and Innovations: 16th IFIP 23 WG 12.5 International Conference, AIAI 2020, Neos Marmaras, Greece, June 5–7, 2020, Proceedings, Part II 16, Springer, 202...

work page 2020

[10] [10]

Yang, X.-Y

L. Yang, X.-Y . Liu, X. Li, Y . Li, Price prediction of cryptocurrency: an empirical study, in: Smart Blockchain: Second International Conference, SmartBlock 2019, Birmingham, UK, October 11–13, 2019, Proceedings 2, Springer, 2019, pp. 130–139

work page 2019

[11] [11]

A. M. Khedr, I. Arif, M. El-Bannany, S. M. Alhashmi, M. Sreedharan, Cryptocurrency price prediction using traditional statistical and machine-learning techniques: A survey, Intelligent Systems in Accounting, Finance and Management 28 (1) (2021) 3–34

work page 2021

[12] [12]

M. M. Patel, S. Tanwar, R. Gupta, N. Kumar, A deep learning-based cryptocurrency price prediction scheme for financial institutions, Journal of information security and applications 55 (2020) 102583

work page 2020

[13] [13]

Vikram, N

K. Vikram, N. Sivaraman, P. Balamurugan, Crypto currency market price prediction using data science process, International Journal for Research in Applied Science and Engineering Technology 10 (2) (2022) 1451–1454

work page 2022

[14] [14]

Deprez, M

N. Deprez, M. Frömmel, Are simple technical trading rules profitable in bitcoin markets?, International Review of Economics & Finance 93 (2024) 858–874

work page 2024

[15] [15]

W. N. Goetzmann, A. Kumar, Equity portfolio diversification, Review of Finance 12 (3) (2008) 433–463

work page 2008

[16] [16]

Basak, A

S. Basak, A. Pavlova, A. Shapiro, Optimal asset allocation and risk shifting in money management, The Review of Financial Studies 20 (5) (2007) 1583–1621

work page 2007

[17] [17]

Brauneis, R

A. Brauneis, R. Mestel, Cryptocurrency-portfolios in a mean-variance framework, Finance Research Letters 28 (2019) 259–264

work page 2019

[18] [18]

F. J. Fabozzi, H. M. Markowitz, F. Gupta, Portfolio selection, Handbook of Finance 2 (2008)

work page 2008

[19] [19]

Chaves, J

D. Chaves, J. Hsu, F. Li, O. Shakernia, Risk parity portfolio vs. other asset allocation heuristic portfolios, Journal of Investing 20 (1) (2011) 108

work page 2011

[20] [20]

Amirzadeh, A

R. Amirzadeh, A. Nazari, D. Thiruvady, Applying artificial intelligence in cryptocurrency markets: A survey, Algorithms 15 (11) (2022) 428

work page 2022

[21] [21]

Jiang, J

Z. Jiang, J. Liang, Cryptocurrency portfolio management with deep reinforcement learning, in: 2017 Intelligent systems conference (IntelliSys), IEEE, 2017, pp. 905–913. 24

work page 2017

[22] [22]

Karalevicius, N

V . Karalevicius, N. Degrande, J. De Weerdt, Using sentiment analysis to predict interday bitcoin price move- ments, The Journal of Risk Finance 19 (1) (2018) 56–75

work page 2018

[23] [23]

Kim, S.-Y

K. Kim, S.-Y . T. Lee, S. Assar, The dynamics of cryptocurrency market behavior: sentiment analysis using markov chains, Industrial Management & Data Systems 122 (2) (2022) 365–395

work page 2022

[24] [24]

P. Rao, N. Bhasin, P. Goswami, L. Aggarwal, Crypto currency portfolio allocation using machine learning, in: 2021 3rd International Conference on Advances in Computing, Communication Control and Networking (ICAC3N), IEEE, 2021, pp. 1522–1527

work page 2021

[25] [25]

E. W. V . Zuniga, C. M. Ranieri, L. Zhao, J. Ueyama, Y .-t. Zhu, D. Ji, Maximizing portfolio profitability during a cryptocurrency downtrend: A bitcoin blockchain transaction-based approach, Procedia Computer Science 222 (2023) 539–548

work page 2023

[26] [26]

T. Lan, M. Frömmel, Risk factors in cryptocurrency pricing, International Review of Financial Analysis (2025) 104389

work page 2025

[27] [27]

M. Y . Hong, J. W. Yoon, The impact of covid-19 on cryptocurrency markets: A network analysis based on mutual information, Plos ONE 17 (2) (2022) e0259869

work page 2022

[28] [28]

A. Rea, W. Rea, Visualization of a stock market correlation matrix, Physica A: Statistical Mechanics and its Applications 400 (2014) 109–123

work page 2014

[29] [29]

Lyócsa, T

Š. Lyócsa, T. V `yrost, E. Baumöhl, Stock market networks: The dynamic conditional correlation approach, Physica A: Statistical Mechanics and its Applications 391 (16) (2012) 4147–4158

work page 2012

[30] [30]

Onnela, A

J.-P. Onnela, A. Chakraborti, K. Kaski, J. Kertesz, A. Kanto, Dynamics of market correlations: Taxonomy and portfolio analysis, Physical Review E 68 (5) (2003) 056110

work page 2003

[31] [31]

R. Jing, L. E. Rocha, A network-based strategy of price correlations for optimal cryptocurrency portfolios, Finance Research Letters 58 (2023) 104503

work page 2023

[32] [32]

Ioannidis, I

E. Ioannidis, I. Sarikeisoglou, G. Angelidis, Portfolio construction: A network approach, Mathematics 11 (22) (2023) 4670

work page 2023

[33] [33]

Tumminello, T

M. Tumminello, T. Aste, T. Di Matteo, R. N. Mantegna, A tool for filtering information in complex systems, Proceedings of the National Academy of Sciences 102 (30) (2005) 10421–10426. 25

work page 2005

[34] [34]

A. J. McNeil, R. Frey, P. Embrechts, Quantitative risk management: concepts, techniques and tools-revised edition, Princeton university press, 2015

work page 2015

[35] [35]

P. J. Kaufman, Trading systems and methods, V ol. 591, John Wiley & Sons, 2013

work page 2013

[36] [36]

Pardo, The evaluation and optimization of trading strategies, John Wiley & Sons, 2011

R. Pardo, The evaluation and optimization of trading strategies, John Wiley & Sons, 2011

work page 2011

[37] [37]

V . D. Blondel, J.-L. Guillaume, R. Lambiotte, E. Lefebvre, Fast unfolding of communities in large networks, Journal of statistical mechanics: theory and experiment 2008 (10) (2008) P10008

work page 2008

[38] [38]

Lancichinetti, S

A. Lancichinetti, S. Fortunato, Consensus clustering in complex networks, Scientific reports 2 (1) (2012) 336

work page 2012

[39] [39]

Akaike, A new look at the statistical model identification, IEEE transactions on automatic control 19 (6) (1974) 716–723

H. Akaike, A new look at the statistical model identification, IEEE transactions on automatic control 19 (6) (1974) 716–723

work page 1974

[40] [40]

Abadi, P

M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, M. Devin, S. Ghemawat, G. Irving, M. Isard, et al., Tensorflow: a system for large-scale machine learning, in: 12th USENIX symposium on operating systems design and implementation (OSDI 16), 2016, pp. 265–283

work page 2016

[41] [41]

O. B. Sezer, M. U. Gudelek, A. M. Ozbayoglu, Financial time series forecasting with deep learning: A systematic literature review: 2005–2019, Applied soft computing 90 (2020) 106181

work page 2005

[42] [42]

L. E. Rocha, N. Masuda, P. Holme, Sampling of temporal networks: Methods and biases, Physical Review E 96 (5) (2017) 052302

work page 2017

[43] [43]

R. N. Mantegna, Hierarchical structure in financial markets, The European Physical Journal B 11 (1) (1999) 193–197

work page 1999

[44] [44]

Pantaleo, M

E. Pantaleo, M. Tumminello, F. Lillo, R. N. Mantegna, When do improved covariance matrix estimators enhance portfolio optimization? an empirical comparative study of nine estimators, Quantitative Finance 11 (7) (2011) 1067–1080

work page 2011

[45] [45]

Ledoit, M

O. Ledoit, M. Wolf, A well-conditioned estimator for large-dimensional covariance matrices, Journal of Multi- variate Analysis 88 (2) (2004) 365–411

work page 2004

[46] [46]

R. A. Martin, Pyportfolioopt: portfolio optimization in python, Journal of Open Source Software 6 (61) (2021) 3066

work page 2021

[47] [47]

Dowd, Measuring market risk, John Wiley & Sons, 2007

K. Dowd, Measuring market risk, John Wiley & Sons, 2007. 26

work page 2007

[48] [48]

V . V . Acharya, L. H. Pedersen, T. Philippon, M. Richardson, Measuring systemic risk, The review of financial studies 30 (1) (2017) 2–47

work page 2017

[49] [49]

Keating, W

C. Keating, W. F. Shadwick, An introduction to omega, AIMA newsletter (2002)

work page 2002

[50] [50]

M. I. Rakib, A. Nobi, J. W. Lee, Structure and dynamics of financial networks by feature ranking method, Scientific Reports 11 (1) (2021) 17618

work page 2021

[51] [51]

T. Aste, T. Di Matteo, M. Tumminello, R. N. Mantegna, Correlation filtering in financial time series, in: Noise and Fluctuations in Econophysics and Finance, V ol. 5848, SPIE, 2005, pp. 100–109. 27 Supplementary Information Data source The daily price data of the cryptocurrencies used in our study was obtained from the following websites: www.investing.co...

work page 2005