Forecasting cryptocurrency returns with machine learning

This article employs machine learning models to predict returns for 3703 cryptocurrencies for the 2013 – 2021 period. Based on daily data, we build an equal (capital)-weighted portfolio that generates 7.1 % (2.4 %) daily return with a 1.95 (0.27) Sharpe ratio. We obtain an out-of-sample R² of 4.855 %. Our results suggest that cryptocurrencies behave like conventional assets than fiat currencies since variables, including lagged returns, can predict future returns. As assets, cryptocurrencies are not weakly efficient, and production costs do not determine their prices. Returns for small cryptocurrencies are more predictable than larger ones. The predictive power of the 1-day lagged return is stronger than all other features (predictors) combined. The results offer new insights for crypto investors, traders, and financial analysts.     

Forecasting stock returns with large dimensional factor models

We study equity premium out-of-sample predictability by extracting the information contained in a high number of macroeconomic predictors via large dimensional factor models. We compare the well-known factor model with a static representation of the common components with the Generalized Dynamic Factor Model, which accounts for time series dependence in the common components. Using statistical and economic evaluation criteria, we empirically show that the Generalized Dynamic Factor Model helps predicting the equity premium. Exploiting the link between business cycle and return predictability, we find accurate predictions also by combining rolling and recursive forecasts in real-time.     

Global integration and emerging stock market excess returns

This article studies the effects of the global integration process on emerging stock market excess returns in a dynamic context. I improve the existing literature in four main directions. First, I show that the average excess returns rise as the level of financial and real integration rises. Second, I find overwhelming evidence that the financial liberalizations (i.e. de jure integration) of the late 1980s and early 1990s have not been simultaneously accompanied by a de facto integration. Third, I find that the percentage of variation in emerging excess returns explained by non-traded global risk factors rises as the level of market openness rises. Last, at the country level, I show that the correlation coefficient does not represent a robust measure of integration. Results also suggest that there are substantial cross-country differences in the dynamics of the degree of financial integration.     

Forecasting stock market returns over multiple time horizons

In this paper, we seek to demonstrate the predictability of stock market returns and explain the nature of this return predictability. To this end, we introduce investors with different investment horizons into the news-driven, analytic, agent-based market model developed in Gusev et al. [Algo. Finance, 2015, 4, 5–51]. This heterogeneous framework enables us to capture dynamics at multiple timescales, expanding the model’s applications and improving precision. We study the heterogeneous model theoretically and empirically to highlight essential mechanisms underlying certain market behaviours, such as transitions between bull and bear markets and the self-similar behaviour of price changes. Most importantly, we apply this model to show that the stock market is nearly efficient on intraday timescales, adjusting quickly to incoming news, but becomes inefficient on longer timescales, where news may have a long-lasting nonlinear impact on dynamics, attributable to a feedback mechanism acting over these horizons. Then, using the model, we design algorithmic strategies that utilize news flow, quantified and measured, as the only input to trade on market return forecasts over multiple horizons, from days to months. The backtested results suggest that the return is predictable to the extent that successful trading strategies can be constructed to harness this predictability.     

Explaining the (non-) adoption of advanced data analytics in auditing: A process theory

Audit firms are increasingly engaging with advanced data analytics to improve the efficiency and effectiveness of external audits through the automation of audit work and obtaining a better understanding of the client’s business risk and thus their own audit risk. This paper examines the process by which audit firms adopt advanced data analytics, which has been left unaddressed by previous research. We derive a process theory from expert interviews which describes the activities within the process and the organizational units involved. It further describes how the adoption process is affected by technological, organizational and environmental contextual factors. Our work contributes to the extent body of research on technology adoption in auditing by using a previously unused theoretical perspective, and contextualizing known factors of technology adoption. The findings presented in this paper emphasize the importance of technological capabilities of audit firms for the adoption of advanced data analytics; technological capabilities within audit teams can be leveraged to support both the ideation of possible use cases for advanced data analytics, as well as the diffusion of solutions into practice.     

Futures volatility forecasting based on big data analytics with incorporating an order imbalance effect

Future markets play vital roles in supporting economic activities in modern society. For example, crude oil and electricity futures markets have heavy effects on a nation’s energy operation management. Thus, volatility forecasting of the futures market is an emerging but increasingly influential field of financial research. In this paper, we adopt big data analytics, called Extreme Gradient Boosting (XGBoost) from computer science, in an attempt to improve the forecasting accuracy of futures volatility and to demonstrate the application of big data analytics in the financial spectrum in terms of volatility forecasting. We further unveil that order imbalance estimation might incorporate abundant information to reflect price jumps and other trading information in the futures market. Including order imbalance information helps our model capture underpinned market rules such as supply and demand, which lightens the information loss during the model formation. Our empirical results suggest that the volatility forecasting accuracy of the XGBoost method considerably beats the GARCH-jump and HAR-jump models in both crude oil futures market and electricity futures market. Our results could also produce plentiful research implications for both policy makers and energy futures market participants.     

Machine learning methods in finance: Recent applications and prospects

We study how researchers can apply machine learning (ML) methods in finance. We first establish that the two major categories of ML (supervised and unsupervised learning) address fundamentally different problems than traditional econometric approaches. Then, we review the current state of research on ML in finance and identify three archetypes of applications: (i) the construction of superior and novel measures, (ii) the reduction of prediction error, and (iii) the extension of the standard econometric toolset. With this taxonomy, we give an outlook on potential future directions for both researchers and practitioners. Our results suggest many benefits of ML methods compared to traditional approaches and indicate that ML holds great potential for future research in finance.     

Non-scheduled news arrival and high-frequency stock market dynamics: Evidence from the Australian Securities Exchange

An increasing number of market participants utilise news analytics software to comprehend the large amounts of unstructured data flowing through news-wires. Utilising original data from one such tool – Ravenpack – I examine the market reaction of leading Australian stocks to stock-specific news flow over an extended period. Unconditional analysis of key variables around 484,440 news items reveals distinct responses in market activity, volatility, bid-ask spreads and returns. The study confirms previous literature such that indicated relevance of news items is critical when identifying significant effects. In addition, the reaction of market activity, volatility and spreads is greatest for negative news. The findings are confirmed when controlling for market dynamics and cross-dependencies between variables in a high-frequency VAR model.     

Evaluating interpretable machine learning predictions for cryptocurrencies

This study explores various machine learning and deep learning applications on financial data modelling, analysis and prediction processes. The main focus is to test the prediction accuracy of cryptocurrency hourly returns and to explore, analyse and showcase the various interpretability features of the ML models. The study considers the six most dominant cryptocurrencies in the market: Bitcoin, Ethereum, Binance Coin, Cardano, Ripple and Litecoin. The experimental settings explore the formation of the corresponding datasets from technical, fundamental and statistical analysis. The paper compares various existing and enhanced algorithms and explains their results, features and limitations. The algorithms include decision trees, random forests and ensemble methods, SVM, neural networks, single and multiple features N-BEATS, ARIMA and Google AutoML. From experimental results, we see that predicting cryptocurrency returns is possible. However, prediction algorithms may not generalise for different assets and markets over long periods. There is no clear winner that satisfies all requirements, and the main choice of algorithm will be tied to the user needs and provided resources.     

Forecasting spot prices of agricultural commodities in India: Application of deep-learning models

Food price fluctuations can impact both producers and consumers. Forecasting the prices of the agricultural commodities is of prime concern not only to the government but also to farmers and agribusiness firms. In developing countries like India, management of food security needs competent and efficient forecasting of food prices. With the availability of data, recent innovation in deep-learning models provides a feasible solution to accurately forecast the prices. In this study, we examine the superiority of these models using the daily spot prices of five major commodities traded on the National Commodity and Derivatives Exchange: cotton seed, castor seed, rape mustard seed, soybean seed, and guar seed. The results were obtained from the application of the traditional univariate autoregressive integrated moving average model and deep-learning techniques like the time-delay neural network (TDNN) and long short-term memory (LSTM) network. The empirical results indicate that the LSTM model is indeed suitable for the financial domain and captures the directional movement of the spot price changes with high accuracy compared with the TDNN and other linear models. Accuracy of the performance of these models has been compared using out-of-sample performance measure. The overall objective of this paper is to demonstrate the utility of spot price forecasting for farmers and traders in offering them the best predictions of the price movements. Our results provide a possibility of developing pricing models that can help in fairly regulating agricultural commodity prices.     

Cointegration, Efficiency and Forecasting in the Currency Market

Existing literature on using the cointegration approach to examine the efficiency of the foreign exchange market gives mixed results. Arguments typically focus on econometric testing techniques, with fractional cointegration being the most current one. This paper tries to look at the issue from an economic perspective. It shows that the cointegrating relationship, whether cointegrated or fractionally cointegrated, is found mainly among the currencies of the European Monetary System which are set to fluctuate within a given range. Hence, there is no inconsistency with the notion of market efficiency. Yet, exploiting such a cointegrating relationship is helpful in currency forecasting. There is some evidence that restricting the forecasting model to consist of only cointegrated currencies improves forecasting efficiency.     

Detecting insurance fraud using supervised and unsupervised machine learning

Fraud is a significant issue for insurance companies, generating much interest in machine learning solutions. Although supervised learning for insurance fraud detection has long been a research focus, unsupervised learning has rarely been studied in this context, and there remains insufficient evidence to guide the choice between these branches of machine learning for insurance fraud detection. Accordingly, this study evaluates supervised and unsupervised learning using proprietary insurance claim data. Furthermore, we conduct a field experiment in cooperation with an insurance company to investigate the performance of each approach in terms of identifying new fraudulent claims. We derive several important findings. Unsupervised learning, especially isolation forests, can successfully detect insurance fraud. Supervised learning also performs strongly, despite few labeled fraud cases. Interestingly, unsupervised and supervised learning detect new fraudulent claims based on different input information. Therefore, for implementation, we suggest understanding supervised and unsupervised methods as complements rather than substitutes.     

Red sky at night or in the morning,to the equity market neither a delight nor a warning: the weather effect re-examined using intraday stock data

Unlike most of the existing literature on the weather effect, we conducted our analysis by employing intraday weather and market data, examining a large set of stocks rather than indices only, including volume and volatility data in the study and inspecting a wide number of weather variables (temperature, humidity, pressure, visibility, wind, cloud, rain and snow). Our analysis covered the Italian stock market for the period August 2005–March 2014 for a total of 2201 trading days. We conclude that no systematic relationship seems to exist between the weather and the Italian stock market. Moreover, our results raise doubts that testing the weather effect by limiting the analysis to indices only can lead to spurious conclusions.     

Real‐Time Forecasting and Political Stock Market Anomalies: Evidence for the United States

Using monthly data from 1953 to 2003, we apply a real‐time modeling approach to investigate the implications of U.S. political stock market anomalies for forecasting excess stock returns in real‐time. Our empirical findings show that political variables, chosen on the basis of widely used model‐selection criteria, are often included in real‐time forecasting models. However, political variables do not contribute systematically to improving the performance of simple trading rules. For this reason, political stock market anomalies are not necessarily an indication of market inefficiency.     

Continuous monitoring with machine learning and interactive data visualization: An application to a healthcare payroll process

This paper presents a framework for proactive and intelligent continuous control monitoring (CCM) that helps management gain higher assurance over business processes and alleviate information overload. We adopt a design science approach towards systematically developing CCM artifacts, including operation and internal control violation display and multidimensional anomaly detection. We illustrate the design with an implementation project whereby a CPA firm, the firm's healthcare sector client, and the research team work together to improve the assurance provided by payroll reviews. This study contributes to the CCM literature by envisioning that interactive data visualization and machine learning technologies can alleviate information overload for management in CCM. Second, we provide real-world evidence on the improvement brought to economic and behavioral aspects of the control monitoring process compared to the traditional approach. We show that emerging technologies substantially improve the efficiency and effectiveness of risk assessment, anomaly detection, and loss prevention. We also contribute to control monitoring practice by providing guidance on artifact development and application for practitioners to follow.     

Encoding of high-frequency order information and prediction of short-term stock price by deep learning

Predicting the price trends of stocks based on deep learning and high-frequency data has been studied intensively in recent years. Especially, the limit order book which describes supply-demand balance of a market is used as the feature of a neural network; however these methods do not utilize the properties of market orders. On the other hand, the order-encoding method of our prior work can take advantage of these properties. In this paper, we apply some types of convolutional neural network architectures to order-based features to predict the direction of mid-price trends. The results show that smoothing filters which we propose to employ rather than embedding features of orders improve accuracy. Furthermore, inspection of the embedding layer and investment simulation are conducted to demonstrate the practicality and effectiveness of our model.     

Empirical analysis of real and financial volatilities on stock excess returns: evidence from Taiwan industrial data

This paper tests the relation between stock excess returns and risk factors measured by volatility. The sources of the volatility are based on the volatility of macroeconomic factors and time-series volatility. To model the macroeconomic fundamentals, we divide the risk into real and financial volatilities pertinent to Taiwan's economic environment. By examining the data of indusry excess returns and market excess returns, we find evidence to reject the hypothesis that the stock excess returns are independent of the real and financial volatilities.     

Twitter Mining for Discovery,Prediction and Causality: Applications and Methodologies

Twitter has found substantial use in a number of settings. For example, Twitter played a major role in the ‘Arab Spring’ and has been adopted by a large number of the Fortune 100. All of these and other events have led to a large database of Twitter tweets that has attracted the attention of a number of companies and researchers through what has become known as ‘Twitter mining’ (also known as ‘TwitterMining’). This paper analyses some of the approaches used to gather information and knowledge from Twitter for Twitter mining. In addition, this paper reviews a number of the applications that employ Twitter Mining, investigating Twitter information for prediction, discovery and as an informational basis of causation. Copyright © 2015 John Wiley & Sons, Ltd.     

Drivers of economic and financial integration: A machine learning approach

We propose a new approach to identifying drivers of economic and financial integration, separately, and across emerging and developed countries. Our advanced machine learning technique allows for nonlinear relationships, corrects for over-fitting, and is less prone to noise. It also can tackle a large number of highly correlated explanatory variables and controls for multicollinearity. Results suggest that general economic growth, increasing international trade, and contained population growth have helped emerging countries catch up to the level of the economic integration of developed countries. However, slow financial development and a high level of investment riskiness have hindered the speed of emerging countries’ financial integration. Furthermore, the results suggest that integration is a gradual process and is not driven by cyclical or transitory events.