Deep convolutional autoencoder for cryptocurrency market analysis

This study attempts to analyze patterns in cryptocurrency markets using a special type of deep neural networks, namely a convolutional autoencoder. The method extracts the dominant features of market behavior and classifies the 40 studied cryptocurrencies into several classes for twelve 6-month periods starting from 15th May 2013. Transitions from one class to another with time are related to the maturement of cryptocurrencies. In speculative cryptocurrency markets, these findings have potential implications for investment and trading strategies.

[1]  Shelton Peiris,et al.  A new look at Cryptocurrencies , 2018 .

[2]  A. H. Dyhrberg.  Hedging capabilities of bitcoin. Is it the virtual gold , 2016 .

[3]  Yulei Rao,et al.  A deep learning framework for financial time series using stacked autoencoders and long-short term memory , 2017, PloS one.

[4]  Sander Bohte,et al.  Conditional Time Series Forecasting with Convolutional Neural Networks , 2017, 1703.04691.

[5]  Seong‐Min Yoon,et al.  Bitcoin as Hedge or Safe Haven: Evidence from Stock, Currency, Bond and Derivatives Markets , 2019, Computational Economics.

[6]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[7]  Sun-Hak Kim,et al.  Using Genetic Algorithm and NARX Neural Network to Forecast Daily Bitcoin Price , 2019, Computational Economics.

[8]  Wang Chun Wei.  Liquidity and market efficiency in cryptocurrencies , 2018, Economics Letters.

[9]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[10]  Jaewook Lee,et al.  An Empirical Study on Modeling and Prediction of Bitcoin Prices With Bayesian Neural Networks Based on Blockchain Information , 2018, IEEE Access.

[11]  Salim Lahmiri,et al.  Cryptocurrency forecasting with deep learning chaotic neural networks , 2019, Chaos, Solitons & Fractals.

[12]  B. LeBaron,et al.  A test for independence based on the correlation dimension , 1996 .

[13]  Timothy Dozat,et al.  Incorporating Nesterov Momentum into Adam , 2016 .

[14]  G. Box,et al.  On a measure of lack of fit in time series models , 1978 .

[15]  J. Wolfowitz,et al.  On a Test Whether Two Samples are from the Same Population , 1940 .

[16]  Qiang Liu,et al.  A Survey of Clustering With Deep Learning: From the Perspective of Network Architecture , 2018, IEEE Access.

[17]  Larisa Yarovaya,et al.  Exploring the Dynamic Relationships between Cryptocurrencies and Other Financial Assets , 2017 .

[18]  Andrew Urquhart.  The Inefficiency of Bitcoin , 2016 .

[19]  Konstantinos Gkillas,et al.  An application of extreme value theory to cryptocurrencies , 2018 .

[20]  D. Baur,et al.  Bitcoin Time-of-Day, Day-of-Week and Month-of-Year Effects in Returns and Trading Volume , 2017, Finance Research Letters.

[21]  Roland Mestel,et al.  Price discovery of cryptocurrencies: Bitcoin and beyond , 2018 .

[22]  Elie Bouri,et al.  Can Volume Predict Bitcoin Returns and Volatility? A Quantiles-Based Approach , 2017 .

[23]  Jürgen Schmidhuber,et al.  Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction , 2011, ICANN.

[24]  A. Ibáñez,et al.  Semi-strong efficiency of Bitcoin , 2018, Finance Research Letters.

[25]  Aviral Kumar Tiwari,et al.  Informational efficiency of Bitcoin—An extension , 2018 .

[26]  A. F. Bariviera.  The Inefficiency of Bitcoin Revisited: A Dynamic Approach , 2017, 1709.08090.

[27]  S. Corbet,et al.  Datestamping the Bitcoin and Ethereum Bubbles , 2017, Finance Research Letters.

[28]  Samir Saadi,et al.  Portfolio diversification with virtual currency: Evidence from bitcoin , 2019, International Review of Financial Analysis.

[29]  En Zhu,et al.  Deep Clustering with Convolutional Autoencoders , 2017, ICONIP.

[30]  Chi Keung Marco Lau,et al.  The causal relationship between Bitcoin attention and Bitcoin returns: Evidence from the Copula-based Granger causality test , 2019, Finance Research Letters.

[31]  Samuel A. Vigne,et al.  Does economic policy uncertainty predict the Bitcoin returns? An empirical investigation , 2018, Finance Research Letters.

[32]  Skipper Seabold,et al.  Statsmodels: Econometric and Statistical Modeling with Python , 2010, SciPy.

[33]  S. Nadarajah,et al.  On the inefficiency of Bitcoin , 2017 .

[34]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[35]  Shane A. Corwin,et al.  A Simple Way to Estimate Bid-Ask Spreads from Daily High and Low Prices , 2011 .

[36]  Nicholas G. Polson,et al.  Deep Learning in Finance , 2016, ArXiv.

[37]  Paraskevi Katsiampa.  Volatility co-movement between Bitcoin and Ether , 2019, Finance Research Letters.

[38]  Cheng Deng,et al.  Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative Entropy Minimization , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[39]  Raghavendra Chalapathy University of Sydney,et al.  Deep Learning for Anomaly Detection: A Survey , 2019, ArXiv.