ACM Transactions on Information Systems · 2009 · 770 citations · 18 references
ComputingComputer ScienceFinanceForecastingMachine LearningPredictive AnalyticsFinancial ForecastingFinancial Time Series AnalysisInformation ExtractionNamed Entity RecognitionSupport Vector MachinesText Feature EngineeringFinancial Time Series ForecastingMachine Learning-based Price ForecastingNews CorporaPredictive Price ModelingStock Market ReturnsSupport Vector Machine ClassificationWord Representation Learning
The study investigates a machine‑learning method that predicts stock price movements from financial news using various textual representations such as bag‑of‑words, noun phrases, and named entities. The authors analyzed 9,211 news articles and over 10 million S&P 500 stock quotes over five weeks, estimating the stock price 20 minutes after each article. An SVM model incorporating article terms and the current stock price achieved the lowest error (MSE 0.04261), 57.1 % directional accuracy, and a 2.06 % simulated return, and a Proper‑Noun representation outperformed Bag‑of‑Words across all metrics.
Our research examines a predictive machine learning approach for financial news articles analysis using several different textual representations: bag of words, noun phrases, and named entities. Through this approach, we investigated 9,211 financial news articles and 10,259,042 stock quotes covering the S&P 500 stocks during a five week period. We applied our analysis to estimate a discrete stock price twenty minutes after a news article was released. Using a support vector machine (SVM) derivative specially tailored for discrete numeric prediction and models containing different stock-specific variables, we show that the model containing both article terms and stock price at the time of article release had the best performance in closeness to the actual future stock price (MSE 0.04261), the same direction of price movement as the future price (57.1% directional accuracy) and the highest return using a simulated trading engine (2.06% return). We further investigated the different textual representations and found that a Proper Noun scheme performs better than the de facto standard of Bag of Words in all three metrics.
18
The Behavior of Stock-Market Prices
Eugene F. Fama · The Journal of Business · 1965 · 8.7K citations
Application of support vector machines in financial time series forecasting
Francis E. H. Tay, Lijuan Cao · Omega · 2001 · 1.3K citations
Computing, Data Science, Finance +14
A hybrid ARIMA and support vector machines model in stock price forecasting
Ping‐Feng Pai, Chih-Sheng Lin · Omega · 2004 · 1K citations
Computing, Finance, Mathematics +16