The Journal of Financial Data Science最新文献

{"title":"Economic Recession Prediction Using Deep Neural Network","authors":"Zihao Wang, Kun Li, Steve Q. Xia, Hongfu Liu","doi":"10.3905/jfds.2022.1.097","DOIUrl":"https://doi.org/10.3905/jfds.2022.1.097","url":null,"abstract":"We investigate the effectiveness of different machine learning methodologies in predicting economic cycles. We identify the deep learning methodology of BiLSTM with autoencoder as the most accurate model to forecast the beginning and end of economic recessions in the United States. We adopt commonly available macro and market-condition features to compare the ability of different machine learning models to generate good predictions both in-sample and out-of-sample. The proposed model is flexible and dynamic when both predictive variables and model coefficients vary over time. It provided good out-of-sample predictions for the past two recessions and early warning about the COVID-19 recession.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121716844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Inside Peek at AI Use in Private Equity","authors":"T. Åstebro","doi":"10.3905/jfds.2021.1.067","DOIUrl":"https://doi.org/10.3905/jfds.2021.1.067","url":null,"abstract":"The number of private equity (PE) firms that have started to use artificial intelligence (AI) in investment decisions has risen rapidly over the past 10 years. This article provides a detailed account that can serve as a template for others in the industry who wish to make better investment decisions using AI. The news is both good and bad. The increased use of AI in PE and venture capital will greatly increase operational efficiency and transform the ways in which partners perform their work. It will allow for the entry of new firms but will also lead to a technological arms race and is predicted to cause an eventual industry shakeout. TOPICS: Private equity, big data/machine learning, performance measurement Key Findings ▪ The number of private equity firms that have started to use AI is rising rapidly. ▪ Use of AI will greatly transform the deal-making process. ▪ By increasing efficiency, AI will likely cause an industry shakeout.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126579278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Context, Language Modeling, and Multimodal Data in Finance","authors":"Sanjiv Ranjan Das, Connor Goggins, John He, G. Karypis, Krishnamurthy Sandeep, Mitali Mahajan, N. Prabhala, Dylan Slack, R. V. Dusen, Shenghua Yue, Sheng Zha, Shuai Zheng","doi":"10.3905/JFDS.2021.1.063","DOIUrl":"https://doi.org/10.3905/JFDS.2021.1.063","url":null,"abstract":"The authors enhance pretrained language models with Securities and Exchange Commission filings data to create better language representations for features used in a predictive model. Specifically, they train RoBERTa class models with additional financial regulatory text, which they denote as a class of RoBERTa-Fin models. Using different datasets, the authors assess whether there is material improvement over models that use only text-based numerical features (e.g., sentiment, readability, polarity), which is the traditional approach adopted in academia and practice. The RoBERTa-Fin models also outperform generic bidirectional encoder representations from transformers (BERT) class models that are not trained with financial text. The improvement in classification accuracy is material, suggesting that full text and context are important in classifying financial documents and that the benefits from the use of mixed data, (i.e., enhancing numerical tabular data with text) are feasible and fruitful in machine learning models in finance. TOPICS: Quantitative methods, big data/machine learning, legal/regulatory/public policy, information providers/credit ratings Key Findings ▪ Machine learning based on multimodal data provides meaningful improvement over models based on numerical data alone. ▪ Context-rich models perform better than context-free models. ▪ Pretrained language models that mix common text and financial text do better than those pretrained on financial text alone.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117107482","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Slow Momentum with Fast Reversion: A Trading Strategy Using Deep Learning and Changepoint Detection","authors":"Kieran Wood, Stephen J. Roberts, S. Zohren","doi":"10.3905/jfds.2021.1.081","DOIUrl":"https://doi.org/10.3905/jfds.2021.1.081","url":null,"abstract":"Momentum strategies are an important part of alternative investments and are at the heart of the work of commodity trading advisors. These strategies have, however, been found to have difficulties adjusting to rapid changes in market conditions, such as during the 2020 market crash. In particular, immediately after momentum turning points, when a trend reverses from an uptrend (downtrend) to a downtrend (uptrend), time-series momentum strategies are prone to making bad bets. To improve the responsiveness to regime change, the authors introduce a novel approach, in which they insert an online changepoint detection (CPD) module into a deep momentum network pipeline, which uses a long short-term memory deep-learning architecture to simultaneously learn both trend estimation and position sizing. Furthermore, their model is able to optimize the way in which it balances (1) a slow momentum strategy that exploits persisting trends but does not overreact to localized price moves and (2) a fast mean-reversion strategy regime by quickly flipping its position and then swapping back again to exploit localized price moves. The CPD module outputs a changepoint location and severity score, allowing the model to learn to respond to varying degrees of disequilibrium, or smaller and more localized changepoints, in a data-driven manner. The authors back test their model over the period 1995–2020, and the addition of the CPD module leads to a 33% improvement in the Sharpe ratio. The module is especially beneficial in periods of significant nonstationarity; in particular, over the most recent years tested (2015–2020), the performance boost is approximately 66%. This is especially interesting because traditional momentum strategies underperformed in this period.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116806670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing Cross-Sectional Currency Strategies by Context-Aware Learning to Rank with Self-Attention","authors":"Daniel Poh, Bryan Lim, S. Zohren, S. Roberts","doi":"10.3905/jfds.2022.1.099","DOIUrl":"https://doi.org/10.3905/jfds.2022.1.099","url":null,"abstract":"The performance of a cross-sectional currency strategy depends crucially on accurately ranking instruments prior to portfolio construction. Although this ranking step is traditionally performed using heuristics or by sorting the outputs produced by pointwise regression or classification techniques, strategies using learning-to-rank algorithms have recently presented themselves as competitive and viable alternatives. Although the rankers at the core of these strategies are learned globally and improve ranking accuracy on average, they ignore the differences between the distributions of asset features over the times when the portfolio is rebalanced. This flaw renders them susceptible to producing suboptimal rankings, possibly at important periods when accuracy is actually needed the most. For example, this might happen during critical risk-off episodes, which consequently exposes the portfolio to substantial, unwanted drawdowns. The authors tackle this shortcoming with an analogous idea from information retrieval: that a query’s top retrieved documents or the local ranking context provide vital information about the query’s own characteristics, which can then be used to refine the initial ranked list. In this work, the authors use a context-aware learning-to-rank model that is based on the transformer architecture to encode top/bottom-ranked assets, learn the context and exploit this information to rerank the initial results. Back testing on a slate of 31 currencies, the authors’ proposed methodology increases the Sharpe ratio by around 30% and significantly enhances various performance metrics. Additionally, this approach also improves the Sharpe ratio when separately conditioning on normal and risk-off market states.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128006592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Managing Editor’s Letter","authors":"F. Fabozzi","doi":"10.3905/jfds.2021.3.2.001","DOIUrl":"https://doi.org/10.3905/jfds.2021.3.2.001","url":null,"abstract":"","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127741683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Promises and Pitfalls of Machine Learning for Predicting Stock Returns","authors":"E. Leung, Harald Lohre, David Mischlich, Yifei Shea, Maximilian Stroh","doi":"10.2139/ssrn.3546725","DOIUrl":"https://doi.org/10.2139/ssrn.3546725","url":null,"abstract":"Recent research suggests that machine learning models dominate traditional linear models in predicting cross-sectional stock returns. The authors confirm this finding when predicting one-month-forward-looking returns based on a set of common stock characteristics, including predictors such as short-term reversal. Despite the statistical advantage of machine learning model predictions, the authors demonstrate that the economic gains tend to be more limited and critically dependent on the ability to take risk and implement trades efficiently. Unlike traditional models, machine learning models have been somewhat more effective over the past decade at discerning valuable predictions from cross-sectional equity characteristics. TOPICS: Security analysis and valuation, big data/machine learning Key Findings ▪ The authors compare a nonlinear machine learning model called gradient boosting machine (GBM) with traditional linear models in predicting cross-sectional stock returns based on well-known equity characteristics. ▪ They demonstrate how to rationalize the mechanics and outcome of GBM to alleviate its black-box characteristics. ▪ The extent to which the statistical advantage of GBM’s performance over that of linear models can be translated into economic gains depends critically on one’s ability to take risk and implement trades efficiently.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117038811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Online Learning with Radial Basis Function Networks","authors":"Gabriel Borrageiro, Nikan B. Firoozye, P. Barucca","doi":"10.36227/techrxiv.14851077.v1","DOIUrl":"https://doi.org/10.36227/techrxiv.14851077.v1","url":null,"abstract":"The authors provide multi-horizon forecasts on the returns of financial time series. Their sequentially optimised radial basis function network (RBFNet) outperforms a random-walk baseline and several powerful supervised learners. Their RBFNets naturally measure the similarity between test samples and prototypes that capture the characteristics of the feature space. The authors show that the training set financial time series returns have low similarity with their test set counterparts, highlighting the challenges faced in particular by kernel-based methods that use the training set returns as test-time prototypes; in contrast, their online learning RBFNets have hidden units that retain greater similarity across time.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123366209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Managing Editor’s Letter","authors":"F. Fabozzi","doi":"10.3905/jfds.2021.3.1.001","DOIUrl":"https://doi.org/10.3905/jfds.2021.3.1.001","url":null,"abstract":"Cathy Scott General Manager and Publisher In implementing a strategic asset allocation policy, the forecasting of long-term equity market returns is critically important. Although, historically, several econometric models have been employed for forecasting, more recently machine learning methods have been used for that purpose. In “The Best of Both Worlds: Forecasting US Equity Market Returns Using a Hybrid Machine Learning–Time Series Approach,” Haifeng Wang, Harshdeep Singh Ahluwalia, Roger A. Aliaga-Díaz, and Joseph H. Davis explore machine learning methods to forecast 10-year-ahead US stock returns. To compare the relative performance of machine learning methods, the authors compare the accuracy of these methods to the forecasts of one of the most commonly used regression-based forecasting models by asset managers, the traditional Shiller cyclically adjusted price-to-earnings (CAPE) ratio model. The authors find that machine learning techniques can only modestly improve the forecast accuracy of the regression-based CAPE ratio model. Moreover, they actually result in worse performance than a vector autoregressive model (VAR)–based two-step approach introduced in 2018 by three of the authors of this article. However, when the authors implement a hybrid ML-VAR approach (i.e., VAR-based two-step approach with machine learning techniques allowing for unspecified nonlinear relationships), they find up to 56% improvement in real-time forecast accuracy for 10-year annualized US stock returns. They find the ensemble method consistently offers the best out-of-sample forecast. Machine learning applications in finance have shown benefits over traditional linear models in forecasting stock returns. Edward Leung, Harald Lohre, David Mischlich, Yifei Shea, and Maximilian Stroh quantify these benefits by comparing the forecasting performance of commonly used machine learning algorithms with that of traditional linear methods in their article “The Promises and Pitfalls of Machine Learning for Predicting Stock Returns.” Using well-known equity characteristics, the authors forecast returns for largeand mid-cap stocks from various regional indexes using a gradient boosting machine (GBM) algorithm and standard ordinary least squares (OLS) approaches. In doing so, they shed light on the mechanics and results of the GBM model in order to alleviate its black-box character. While the forecasts from GBM models outperform OLS models based on statistical tests of forecasting performance, the economic gains from such nonlinear models depend on the ability to take the appropriate risks and efficiently implement trades. Jochen Papenbrock, Peter Schwendner, Markus Jaeger, and Stephan Krügel in their article “Matrix Evolutions: Synthetic Correlations and Explainable Machine Learning for Constructing Robust Investment Portfolios” use evolutionary algorithms to simulate correlation matrixes useful for financial market applications. Referring to their novel approach to generate","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130865385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Reinforcement Learning with Function Properties in Mean Reversion Strategies","authors":"Sophia Gu","doi":"10.3905/fjds.2022.1.094","DOIUrl":"https://doi.org/10.3905/fjds.2022.1.094","url":null,"abstract":"Over the past decades, researchers have been pushing the limits of deep reinforcement learning (DRL). Although DRL has attracted substantial interest from practitioners, many are blocked by having to search through a plethora of available methodologies that are seemingly alike, whereas others are still building RL agents from scratch based on classical theories. To address the aforementioned gaps in adopting the latest DRL methods, the author is particularly interested in testing out whether any of the recent technology developed by the leads in the field can be readily applied to a class of optimal trading problems. Unsurprisingly, many prominent breakthroughs in DRL are investigated and tested on strategic games—from AlphaGo to AlphaStar and, at about the same time, OpenAI Five. Thus, in this article, the author shows precisely how to use a DRL library that is initially built for games in a commonly used trading strategy—mean reversion. And by introducing a framework that incorporates economically motivated function properties, they also demonstrate, through the library, a highly performant and convergent DRL solution to decision-making financial problems in general.","PeriodicalId":199045,"journal":{"name":"The Journal of Financial Data Science","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133322854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}