Predicting Risk from Financial Reports with Regression

By Shimon Kogan, Dimitry Levin, Bryan R. Routledge, Jacob S. Sagi, Noah A. Smith. North American Chapter of the Association for Computational Linguistics, 2009

328 citations recorded by the source corpus.

We address a text regression problem: given a piece of text, predict a real-world continuous quantity associated with the text's meaning. In this work, the text is an SEC-mandated financial report published annually by a publicly-traded company, and the quantity to be predicted is volatility of stock returns, an empirical measure of financial risk. We apply well-known regression techniques to a large corpus of freely available financial reports, constructing regression models of volatility for the period following a report. Our models rival past volatility (a strong baseline) in predicting the target variable, and a single model that uses both can significantly outperform past volatility. Interestingly, our approach is more accurate for reports after the passage of the Sarbanes-Oxley Act of 2002, giving some evidence for the success of that legislation in making financial reports more informative.

Subfields: Accounting, Finance, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/8c2d897f005bfe8e9db94bde269d4eb8b0230f78

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Accounting

In Finance

In Management Science and Operations Research

Citations in the corpus

On the Importance of Text Analysis for Stock Price Prediction (2014, methodology): “We also tried bigram and dependency-based features which were shown to be effective for similar tasks in previous works (Wang and Manning, 2012; Kogan et al., 2009; Engelberg, 2008).”
Deep Learning for Event-Driven Stock Prediction (2015, neutral): “Pioneering work mainly uses simple features from news documents, such as bags-of-words, noun phrases, and named entities [Kogan et al., 2009; Schumaker and Chen, 2009].”
Predicting In-game Actions From the Language of NBA Players (2019, methodology): “a text-as-data, has led to many advancements in the prediction of stock prices (Kogan et al. 2009), understanding of political discourse (Field”
GraphLab: A New Framework For Parallel Machine Learning (2010, methodology): “We evaluate the performance of the GraphLab implementation on a financial data set obtained from Kogan et al. [2009]. The task is to use word counts of a financial report to predict stock volatility of the issuing company for the consequent 12 months.”
Frank-Wolfe with Subsampling Oracle (2018, methodology): “On ﬁgure 3, we test the Lasso problem on the E2006-tf-idf data set (Kogan et al., 2009), which gathers volatility of stock returns from companies with ﬁnancial reports.”
Linguistic Structured Sparsity in Text Categorization (2014, background): “Forecasting from text requires identifying textual correlates of a response variable revealed in the future, most of which will be weak and many of which will be spurious (Kogan et al., 2009).”
Maximin effects in inhomogeneous large-scale data (2014, methodology): “Kogan et al. (2009) collected a dataset of so-called 10-”
Stochastic Alternating Direction Method of Multipliers (2013, methodology): “The features are tf-idf of unigrams extracted from the financial reports of companies during the years 1996-2006 (Kogan et al., 2009).”
Making the Most of Bag of Words: Sentence Regularization with Alternating Direction Method of Multipliers (2014, background): “Forecasting from text requires identifying textual correlates of a response variable revealed in the future, most of which will be weak and many of which will be spurious (Kogan et al., 2009).”
Measuring the Information Content of Financial News (2016, background): “In terms of corpora, ﬁnancial news (Leinweber and Sisk, 2011; Xie et al., 2013; Luss and d’Aspremont, 2015; Ding et al., 2015), ﬁrm reports (Kogan et al., 2009; Li, 2010; Lee et al., 2014; Qiu et al., 2014) and web content, such as tweets (Bollen et al., 2011; Vu et al., 2012) and forum posts (Das…”
Parallel Coordinate Descent for L1-Regularized Loss Minimization (2011, background): “Large, Sparse Datasets: Very large and sparse problems, including predicting stock volatility from text in financial reports (Kogan et al., 2009).”
A Semiparametric Gaussian Copula Regression Model for Predicting Financial Risks from Earnings Calls (2014, methodology): “Kogan et al. (2009) model the SEC-mandated annual reports, and performs linear SVM regression with -insensitive loss function to predict the measured volatility.”
Predicting abnormal returns from news using text classification (2008, methodology): “Kogan et al. (2009) use Support Vector Regression (SVR) to forecast stock return volatility based on text in SEC mandated 10-K reports.”
Semantic Frames to Predict Stock Price Movement (2013, background): “Kogan et al. (2009) address a text regression problem to predict the financial risk of investment in companies.”