1 / 10

News Analytics: Models that Quantify News By Armando Gonzalez President & CEO – RavenPack

News Analytics: Models that Quantify News By Armando Gonzalez President & CEO – RavenPack July 2, 2008. 2. News Information Overload. Analysts and traders are overloaded with news. Increasing amounts of news providers and distribution channels

flynn
Download Presentation

News Analytics: Models that Quantify News By Armando Gonzalez President & CEO – RavenPack

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. News Analytics: Models that Quantify News By Armando Gonzalez President & CEO – RavenPack July 2, 2008

  2. 2 News Information Overload • Analysts and traders are overloaded with news • Increasing amounts of news providers and distribution channels • Newswires (Dow Jones, ThomsonReuters, Bloomberg) • Financial Sites (Marketwatch, Forbes, WSJ, etc.) • Social Media (Blogs, Message Boards, Forums, etc.) • Searching media is limited to keywords and more advanced searches are too complex and time consuming • Detecting new trends and opportunities in news is difficult to discover until the headlines make it to the Front Page • Reading and interpreting news from ALL available sources is time consuming and simply impossible for analysts and traders.

  3. 3 Challenges Incorporating News in Trading There are five major challenges facing a trading firm when incorporating large amounts of news information: • Getting news in a machine-readable format (MRN) • Minimizing News Delivery Latency • Have access to historical news data for backtesting • Access the best tools to handle and manipulate news data • Derive valuable analytics from MRN and apply them in a profitable way

  4. 4 Quantifying News There are various ways to quantify aspects of news stories: • Measure news volume about a specific entity or topic (i.e., count the number of stories about Yahoo or Sub-prime) • Attribute quantifiable properties to news articles (i.e. source quality, relevance, novelty, etc.) • Derive time series representations of news properties • Examine linguistic style (i.e., positive or negative, optimistic or pessimistic tone) • Calculate relationships or correlations between news properties and market prices and trading volume

  5. 5 News Data Properties Sentiment POS: 0.490 OPT: 0.820 Compression Relevance 72.54% MSFT: 0.345 1 = new/latest Information Density 0.8493 Novelty 009 = PRESS RELEASE 9/10 Source Quality News Type

  6. 6 Models that Quantify News Beyond the Numbers: Managers' Use of Optimistic and Pessimistic Tone in Earnings Press Releases – A.K. Davis, J. Piger, & L. Sedor 2007 • Examine whether managers use linguistic style (i.e., optimistic and pessimistic tone) in earnings press releases and how the market responds • Measure tone for approximately 23,400 earnings press releases issued between 1998 and 2003 • Find a significant positive (negative) association between levels of optimistic (pessimistic) tone in earnings press releases and future ROA • Results suggest that managers use optimistic and pessimistic tone in earnings press releases to provide investors with information about expected future firm performance and that the market responds to these disclosures Download: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=875399

  7. 7 Models that Quantify News Giving Content to Investor Sentiment: The Role of Media in the Stock Market – Paul Tetlock 2007 • Explores the interactions between media content and stock market activity • Quantitatively measures the nature of the media’s interactions with the stock market using daily content from a popular Wall Street Journal column • Finds that high media pessimism predicts downward pressure on market prices followed by a reversion to fundamentals • Finds unusually high or low pessimism predicts high market trading volume Download: http://www.mccombs.utexas.edu/faculty/paul.tetlock/papers/Tetlock_Media_Sentiment_JF.pdf

  8. 8 Impact: Negative sentiment shock on prices Source: Giving Content to Investor Sentiment: The Role of Media in the Stock Market – P. Tetlock 2007

  9. 9 Models that Quantify News Quantifying News Sentiment – G. Melis 2008 • Market sentiment is given a quantitative interpretation • Defines sentiment solely on news completely disregarding direct market information • Measures sentiment of news stories and demonstrates significant correlations with daily S&P 500 returns • Experimental findings suggest that, despite electronic trading, at market open stock prices on the whole play catch up incorporating relatively old news • Based on a news sentiment signal a basic long/short trading strategy is shown to outperform the market in five consecutive six month periods Source: RavenPack 2008

  10. 10 Concluding Remarks • Firms are overloaded with information and have turned to computers to read news and internet information • With new technologies and research, Trading Firms are learning to react much faster to ever-increasing amounts of news and information available for making decisions • More and more studies show how news analytics can enhance a firms’ trading strategies, help them better manage risk, and even generate alpha

More Related