Predicting the risk of death for cryptocurrencies

Year 2022, Volume: 8 Issue: 3, 548 - 565, 30.10.2022

Hulya Ozuysal Murat Atan H. Altay Güvenir

https://doi.org/10.30855/gjeb.2022.8.3.011

Abstract

The increasing popularity of cryptocurrencies in recent years has managed to attract the attention of investors. Investors, evaluating their investments with cryptocurrencies, which are speculative investment tools, invest in these currencies with very high volatility. However, many of the cryptocurrencies, whose numbers have increased rapidly in recent years and reached thousands, die before completing even a one-year time frame. This creates a serious societal impact that causes investors to lose significant amount of money. This article examined 2.825 cryptocurrencies, which started to be traded in 2017 and later. The article proposes a number of models that can be used to predict market risk for a portfolio of cryptocurrencies. Model is a methodology for ranking the risk of death for cryptocurrencies using only market closing prices and total daily volume. For this purpose, simple recurrent neural networks, a supervised machine learning method, are used to predict the death for cryptocurrencies. Our models rank the risk of dying in the next 30, 60, 90, 120, and 150 days using the retrospective 30-day performance of cryptocurrencies. As such, the models will be able to serve as a screening tool for investors looking to improve overall portfolio performance and avoid investing in high-risk cryptocurrencies. The article also contributes to the literature on the use of machine learning techniques in calculating the risk of death for cryptocurrencies. In the study, the best performance with the simple recurrent neural network model was obtained in Scenario 5 with a rate of 72.24% AUC. With this scenario, the probability of predicting a dead cryptocurrency as dead is 83.74%. From a financial point of view, it can be suggested as an acceptable value to be able to reduce the probability of failure of the investment by about eighty-four percent.

Keywords

cryptocurrency, dead cryptocurrency, credit risk, recurrent neural network

Kripto para birimlerinin ölme riskinin tahmini

Abstract

Son yıllarda kripto paraların artan popülaritesi yatırımcıların da dikkatini çekmeyi başarmıştır. Yatırımlarını spekülatif bir yatırım aracı olan kripto paralarla da değerlendirmek isteyen yatırımcılar volatilesi çok yüksek olan bu paralara yatırım yapmaktadır. Ancak son yıllarda sayıları hızla artan binlere ulaşan kripto para birimlerinin birçoğu bir yıllık zaman dilimini bile tamamlayamadan ölmektedir. Bu durum yatırımcıların önemli miktarda para kaybetmesine neden olan ciddi bir toplumsal etki yaratmaktadır. Bu makalede, 2017 ve sonrasında arz edilerek işlem görmeye başlayan ve sonrasında ölü olarak kabul edilen 2.825 kripto para birimi incelenmiştir. Makale, bir kripto para portföyü için piyasa riskini tahmin etmek için kullanılabilecek bir dizi model önermektedir. Modeller, kripto paraların yalnızca piyasa kapanış fiyatlarını ve günlük toplam hacimlerini kullanarak ölme riski sıralamasını yapmaya yönelik bir yöntem bilim önerisidir. Bu amaçla ölecek olan kripto paraları tahmin etmek için denetimli bir makine öğrenmesi yöntemi olan basit tekrarlayan sinir ağları kullanılmıştır. Modeller kripto paraların geriye dönük 30 günlük performanslarını kullanarak gelecek 30, 60, 90, 120 ve 150 gün içinde ölme riskini sıralamaktadır. Böylelikle, modeller, genel portföy performansını artıracak ve yüksek riskli kripto para birimlerine yatırım yapmaktan kaçınmak isteyen yatırımcılar için bir tarama aracı olarak hizmet edebilecektir. Makale ayrıca kripto paraların ölme riskinin hesaplanmasında makine öğrenmesi tekniklerinin kullanımı konusunda alan yazına katkıda da bulunmaktadır. Çalışmada basit tekrarlayan sinir ağı modeli ile en iyi performans % 72,24 AUC oranı ile Senaryo 5’de elde edilmiştir. Bu senaryo ile ölen bir kripto paranın ölü olarak tahmin edilme olasılığı % 83,74’dür. Finansal açıdan yaklaşık yüzde seksen dört oranında yatırımın başarısız olma olasılığını azaltabilmek kabul edilebilir bir değer olarak önerilebilir.

Keywords

kripto para, ölü kripto para, kredi riski, tekrarlayan sinir ağları

References

• Bengio, Y., Simard, P., ve Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE transactions on neural networks, 5(2), 157-166.
• Bouoiyour, J., ve Selmi, R. (2015). What does Bitcoin look like? Annals of Economics ve Finance, 16(2).
• Burniske, C., ve White, A. (2017). Bitcoin: Ringing the bell for a new asset class. Ark Invest (January 2017) https://research. ark-invest. com/hubfs/1_Download_Files_ARK-Invest/White_Papers/Bitcoin-Ringing-The-Bell-For-A-New-Asset-Class. pdf.
• Çarkacı, N. (2018). Derin Öğrenme Uygulamalarında En Sık kullanılan Hiper-parametreler. Retrieved from https://medium.com/deep-learning-turkiye/derin-ogrenme-uygulamalarinda-en-sik-kullanilan-hiper-parametreler-ece8e9125c4
• Dowd, K. (2014). New private monies: A bit-part player? Institute of Economic Affairs Monographs, Hobart Paper, 174.
• Fang, F., Ventre, C., Basios, M., Kanthan, L., Martinez-Rego, D., Wu, F., ve Li, L. (2022). Cryptocurrency trading: a comprehensive survey. Financial Innovation, 8(1), 1-59.
• Fantazzini, D., ve Zimin, S. (2019). A multivariate approach for the simultaneous modelling of market risk and credit risk for cryptocurrencies. Journal of Industrial and Business Economics, 47(1), 19-69. doi:10.1007/s40812-019-00136-8
• Fawcett, T. (2006). An introduction to ROC analysis. Pattern recognition letters, 27(8), 861-874.
• Foley, S., Karlsen, J. R., ve Putniņš, T. J. (2019). Sex, drugs, and bitcoin: How much illegal activity is financed through cryptocurrencies? The Review of Financial Studies, 32(5), 1798-1853.
• Force, E. C.-A. T. (2019). Crypto-Assets: Implications for financial stability, monetary policy, and payments and market infrastructures. Retrieved from
• Frisby, D. (2014). Bitcoin: the future of money? : Unbound Publishing.
• Fry, J., ve Cheah, E.-T. (2016). Negative bubbles and shocks in cryptocurrency markets. International Review of Financial Analysis, 47, 343-352.
• Glaser, F., Zimmermann, K., Haferkorn, M., Weber, M. C., ve Siering, M. (2014). Bitcoin-asset or currency? revealing users' hidden intentions. Revealing Users' Hidden Intentions (April 15, 2014). ECIS.
• Grobys, K., ve Sapkota, N. (2019). Predicting Cryptocurrency Defaults. SSRN Electronic Journal. doi:10.2139/ssrn.3383535
• Güvenir, H. A., ve Kurtcephe, M. (2012). Ranking instances by maximizing the area under ROC curve. IEEE Transactions on knowledge and Data Engineering, 25(10), 2356-2366.
• Herpel, M. (2010). 2011 Observations on the Digital Currency Industry. Available at SSRN 1721076.
• Hileman, G., ve Rauchs, M. (2017). Global cryptocurrency benchmarking study. Cambridge Centre for Alternative Finance, 33, 33-113. Huang, J., ve Ling, C. X. (2005). Using AUC and accuracy in evaluating learning algorithms. IEEE Transactions on knowledge and Data Engineering, 17(3), 299-310.
• Kamilaris, A., ve Prenafeta-Boldú, F. X. (2018). Deep learning in agriculture: A survey. Computers and electronics in agriculture, 147, 70-90.
• Kartal, E., ve Özen, Z. (2017). Dengesiz veri setlerinde sınıflandırma. Mühendislikte Yapay Zekâ ve Uygulamaları, 1st ed., O. Torkul, S. Gülseçen, Y. Uyaroğlu, G. Çağıl, and MK Uçar, Eds. Sakarya: Sakarya Üniversitesi Kütüphanesi Yayınevi, 109-131.
• Kethineni, S., ve Cao, Y. (2020). The rise in popularity of cryptocurrency and associated criminal activity. International Criminal Justice Review, 30(3), 325-344.
• Kingma, D. P., ve Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980. Kristoufek, L. (2015). What are the main drivers of the Bitcoin price? Evidence from wavelet coherence analysis. PloS one, 10(4), e0123923.
• Maume, P. (2020). Initial coin offerings and EU prospectus disclosure. European Business Law Review, 31(2).
• Murali, J. (2013). A New Coinage. Economic ve Political Weekly, 48(38), 77-78.
• Nakamoto, S. (2008). Bitcoin: A Peer-to-Peer Electronic Cash System. Retrieved from https://bitcoin.org/bitcoin.pdf
• Paul, V., ve Michael, C. (2018). The Age of Crytocurrency: How Bitcoin and Digital Currency are challenging the global economic order. In: CreateSpace Independent Publishing Platform.
• Posner, E. (2014) /Interviewer: A. Nathan. Goldman Sachs Global Investment Research (Vol 21), Top Of Mind
• Raschka, S., ve Mirjalili, V. (2017). Python Machine Learning: Machine Learning and Deep Learning with Python. Scikit-Learn, and TensorFlow. Second edition ed.
• Sid. (2018). How Peng Coin Will Surge 8-12x These Coming Weeks. Retrieved from https://medium.com/@sidbicious123/how-peng-coin-will-surge-8-12x-these-coming-weeks-4026831b31c1
• Tan, H. H., ve Lim, K. H. (2019). Vanishing gradient mitigation with deep learning neural network optimization. Paper presented at the 2019 7th international conference on smart computing ve communications (ICSCC).
• White, L. H. (2015). The market for cryptocurrencies. Cato J., 35, 383.
• Williams, S. (2021). 21 of the Largest Cryptocurrencies Ranked by Investors' Hold Time. Retrieved from https://www.fool.com/investing/2021/12/06/21-cryptocurrencies-ranked-by-investors-hold-time/
• Yermack, D. (2013). Is Bitcoin a real currency? An economic appraisal (No. w19747). National Bureau of Economic Research, 36(2), 843-850.