Financial markets are complex, evolving dynamic systems. Due to their irregularity, financial time series forecasting is regarded as a rather challenging task. In recent years, artificial neural network applications in finance for such tasks as pattern recognition, classification, and time series forecasting have dramatically increased. The objective of this paper is to present this versatile framework and attempt to use it to predict the stock return series of four public-listed companies on the New York Stock Exchange. Our findings coincide with those of Burton Malkiel in his book, A Random Walk Down Wall Street; no conclusive evidence is found that our proposed models can predict the stock return series better than that of a random walk.
The goal of our work was to select a neural network architecture that would give the best prediction of the Bitcoin exchange rate using historical data. Our work fits into the very important topic of predicting the value of the cryptocurrency exchange rate, and makes use of recent data which, as a result of the high Bitcoin exchange rate dynamics of the last year, differs significantly from those of previous years. We propose and test a number of neural network-based architectures and conduct a discussion of the results. Unlike previous state-of-the-art works, we conducted a comprehensive comparison of three different neural network-based models: MLP (multilayer perceptron), LSTM (long short-term memory) and CNN (convolutional neural network). We tested them for a wide range of parameters. The results we present are, to the best of our knowledge, the most up to date when it comes to the application of artificial intelligence methods for the prediction of cryptocurrency exchange rates. The best-performing architectures were used for a website that gives real-time predictions of the Bitcoin exchange rate. The website is available at http://stpbtc-ii.up.krakow.pl/. Source codes of our research are available to download in order to make our experiment reproducible.
Tomáš Šťastný, Jiřà Koudelka, Diana BÃlková, LuboÅ¡ Marek
Cryptocurrencies are a new field of investment opportunities that has experienced a significant growth in the last decade. The crypto market was capitalized at more than USD 3000 bn, having grown from USD 10 m over the period 2011–2021. Generating high returns, investments in cryptocurrencies have also shown high levels of price volatility. By comparing the performance of cryptocurrencies (measured by the crypto index) and standard equities (included in the S&P 500 index), we found that the former has outperformed the latter 14 times over the last two years. In the present paper, we analyzed the 2012–2022 global crypto market developments and main constituents. With a focus on the top 30 cryptocurrencies and their prices, as of 9 April 2022, covering data of the two major market stress events—outbreaks of the COVID-19 pandemic (February 2020) and the Russian invasion of Ukraine (February 2022). We applied the dynamic time warping method including barycentre averaging and k-Shape clustering of time series. The use of the dynamic time warping has been essential for the preparation of data for subsequent clustering and forecasting. In addition, we compared performance of cryptocurrencies and equities. Cryptocurrency time series are rather short, sometimes involving high levels of volatility and including multiple data gaps, whereas equity time series are much longer and well-established. Identifying similarities between them allows analysts to predict crypto prices by considering the evolution of similar equity instruments and their responses to historical events and stress periods. Moreover, we tested various forecasting methods on the 30 cryptocurrencies to compare traditional econometric methods with machine learning approaches.
Since the inception of Bitcoin in 2009, the market of cryptocurrencies has grown beyond initial expectations as daily trades exceed $10 billion. As industries become automated, the need for an automated fraud detector becomes very apparent. Detecting anomalies in real time prevents potential accidents and economic losses. Anomaly detection in multivariate time series data poses a particular challenge because it requires simultaneous consideration of temporal dependencies and relationships between variables. Identifying an anomaly in real time is not an easy task specifically because of the exact anomalistic behavior they observe. Some points may present pointwise global or local anomalistic behavior, while others may be anomalistic due to their frequency or seasonal behavior or due to a change in the trend. In this paper we suggested working on real time series of trades of Ethereum from specific accounts and surveyed a large variety of different algorithms traditional and new. We categorized them according to the strategy and the anomalistic behavior which they search and showed that when bundling them together to different groups, they can prove to be a good real-time detector with an alarm time of no longer than a few seconds and with very high confidence.
Leonardo Kanashiro Felizardo, Francisco Caio Lima Paiva, Catharine de Vita Graves, Élia Yathie Matsumoto · 7 authors
The interdisciplinary relationship between machine learning and financial markets has long been a theme of great interest among both research communities. Recently, reinforcement learning and deep learning methods gained prominence in the active asset trading task, aiming to achieve outstanding performances compared with classical benchmarks, such as the Buy and Hold strategy. This paper explores both the supervised learning and reinforcement learning approaches applied to active asset trading, drawing attention to the benefits of both approaches. This work extends the comparison between the supervised approach and reinforcement learning by using state-of-the-art strategies with both techniques. We propose adopting the ResNet architecture, one of the best deep learning approaches for time series classification, into the ResNet-LSTM actor (RSLSTM-A). We compare RSLSTM-A against classical and recent reinforcement learning techniques, such as recurrent reinforcement learning, deep Q-network, and advantage actor–critic. We simulated a currency exchange market environment with the price time series of the Bitcoin, Litecoin, Ethereum, Monero, Nxt, and Dash cryptocurrencies to run our tests. We show that our approach achieves better overall performance, confirming that supervised learning can outperform reinforcement learning for trading. We also present a graphic representation of the features extracted from the ResNet neural network to identify which type of characteristics each residual block generates.
Since the appearance of Bitcoin, cryptocurrencies have experienced enormousgrowth not only in terms of capitalization but also in number. As a result, thecryptocurrency market can be an attractive arena for investors as it offers manypossibilities, but a difficult one to understand as well. In this work, we aim tosummarize and segment the whole cryptocurrency market in 2018 with the helpof data analysis tools. We will use three different partitional clustering algorithmseach of them using a different representation for cryptocurrencies, namely: yearlymean and standard deviation of the returns, distribution of returns, and timeseries of returns. Since each representation will provide a different andcomplementary perspective of the market, we will also explore the combination ofthe three clustering results to obtain a fine-grained analysis of the main trends ofthe market. Finally, we will analyse the association of the clustering results withother descriptive features of the cryptocurrencies, including the age, technologicalattributes, and financial ratios derived from them. This will help to enhance theprofiling of the clusters with additional insights. As a result, this work offers adescription of the market and a methodology that can be reproduced by investorsthat want to understand the main trends on the market and that look forcryptocurrencies with different financial performance.
Petrônio Cândido de Lima e Silva, Carlos Alberto Severiano, Marcos Antônio Alves, Rodrigo Silva · 6 authors
In this paper we introduce a Non-Stationary Fuzzy Time Series (NSFTS) method with time varying parameters adapted from the distribution of the data. In this approach, we employ Non-Stationary Fuzzy Sets, in which perturbation functions are used to adapt the membership function parameters in the knowledge base in response to statistical changes in the time series. The proposed method is capable of dynamically adapting its fuzzy sets to reflect the changes in the stochastic process based on the residual errors, without the need to retraining the model. This method can handle non-stationary and heteroskedastic data as well as scenarios with concept-drift. The proposed approach allows the model to be trained only once and remain useful long after while keeping reasonable accuracy. The flexibility of the method by means of computational experiments was tested with eight synthetic non-stationary time series data with several kinds of concept drifts, four real market indices (Dow Jones, NASDAQ, SP500 and TAIEX), three real FOREX pairs (EUR-USD, EUR-GBP, GBP-USD), and two real cryptocoins exchange rates (Bitcoin-USD and Ethereum-USD). As competitor models the Time Variant fuzzy time series and the Incremental Ensemble were used, these are two of the major approaches for handling non-stationary data sets. Non-parametric tests are employed to check the significance of the results. The proposed method shows resilience to concept drift, by adapting parameters of the model, while preserving the symbolic structure of the knowledge base.
Businesses whose work hours, work durations & also more compliance records must be trustworthy and verifiable have great difficulties maintaining more accurate & also safe timekeeping. This article looks at a novel approach for smart contract & also blockchain technology integration protection of timekeeping data. Changing from more conventional, centralized time-tracking systems to a distributed ledger creates a safe framework that logs time inputs as unchangeable transactions, therefore preventing any retroactive changes or more unauthorized additions. While smart contracts provide automation of activities like clock-ins, clock-outs & more compliance alerts, every input is validated & also documented with cryptographic certainty. This guarantees real-time policy compliance and eliminates errors in hand-made monitoring. Clear audit trail, real-time notifications for errors or transgressions, and tamper-evident recordkeeping define the main characteristics of the recommended solution. These elements taken together provide a mechanism wherein auditors, managers, and employees may trust the data free from middlemen's intervention. Extended data integrity, increased organizational trust, greater transparency in payroll and compliance procedures, and fewer incidence of fraud or human error define the projected benefits. From manufacturing to logistics, remote work, healthcare, legal compliance management, the approach is versatile and scalable across numerous sectors. This work integrates trust into the infrastructure via blockchain and smart contract automation therefore establishing a progressive base for future timekeeping systems
Long-term prediction of multivariate time series is still an important but challenging problem. The key to solve this problem is to capture the spatial correlations at the same time, the spatio-temporal relationships at different times and the long-term dependence of the temporal relationships between different series. Attention-based recurrent neural networks (RNN) can effectively represent the dynamic spatio-temporal relationships between exogenous series and target series, but it only performs well in one-step time prediction and short-term time prediction. In this paper, inspired by human attention mechanism including the dual-stage two-phase (DSTP) model and the influence mechanism of target information and non-target information, we propose DSTP-based RNN (DSTP-RNN) and DSTP-RNN-2 respectively for long-term time series prediction. Specifically, we first propose the DSTP-based structure to enhance the spatial correlations between exogenous series. The first phase produces violent but decentralized response weight, while the second phase leads to stationary and concentrated response weight. Secondly, we employ multiple attentions on target series to boost the long-term dependence. Finally, we study the performance of deep spatial attention mechanism and provide experiment and interpretation. Our methods outperform nine baseline methods on four datasets in the fields of energy, finance, environment and medicine, respectively.
Cryptocurrencies are known as unpredictable due to their highly volatility. In time series, the forecasting accuracy is strongly affected by the methodologies that are used in identifying the pattern of a nonstationary stochastic realization. The purpose of the present study is to develop an algorithm that is capable of efficiently identifying the pattern of cryptocurrencies. A brief summary of the algorithm is given. To illustrate the quality of our proposed algorithm, we study the pattern of ten different reputable cryptocurrencies and use their daily closing prices to constitute a time series. The comparison between our proposed forecasting algorithm versus the autoregressive integrated moving average (ARIMA) process will be demonstrated.