Jump to content

AZFinText: Difference between revisions

Add links
From Wikipedia, the free encyclopedia
Browse history interactively
Content deleted Content added
VisualWikitext
DBCubix (talk | contribs)
92 edits
←Created page with '{{Userspace draft|source=ArticleWizard|date={{Subst:CURRENTMONTHNAME}} {{Subst:CURRENTYEAR}}}} {{Subst:Nul|<==do not change this line it will set the date automatic...'
 
AnnaBSP (talk | contribs)
252 edits
No edit summary
 
(32 intermediate revisions by 21 users not shown)
Line 1: Line 1:
{{coi|date=February 2022}}
{{Userspace draft|source=ArticleWizard|date=June 2010}}
'''Arizona Financial Text System''' ('''AZFinText''') is a textual-based quantitative financial prediction system written by [[Robert P. Schumaker]] of [[University of Texas at Tyler]] and [[Hsinchun Chen]] of the [[University of Arizona]].


==System==
'''Arizona Financial Text System (AZFinText)''' is a quantitative textual financial prediction system written by Robert P. Schumaker of [[Iona College]] and Hsinchun Chen of the [[University_Of_Arizona|University of Arizona]]. This system differs from other quants because it uses financial text as one of its key means of predicting stock price movement. This reduces the information lag-time problem evident in many similar systems where new information must be transcribed (e.g., such as losing a costly court battle or having a product recall), before the quant can react appropriately. AZFinText overcomes these limitations by utilizing the terms used in financial news articles to predict future stock prices twenty minutes after the news article has been released <ref>Schumaker, R., (2006). [http://www.robschumaker.com/publications/AMCIS%20-%20Textual%20Analysis%20of%20Stock%20Market%20Prediction%20Using%20Financial%20News%20Articles.pdf Textual Analysis of Stock Market Prediction Using Financial News Articles]. ''12th Americas Conference on Information Systems (AMCIS-2006)'', Aug 2006. Acapulco, Mexico.</ref>. It is believed that certain article terms can move stocks more than others. Terms such as ''factory exploded'' or ''workers strike'' will have a depressing effect on stock prices whereas terms such as ''earnings rose'' will tend to increase stock prices. When a human trading expert sees certain terms, they will react in a somewhat predictable fashion. AZFinText capitalizes on the arbitrage opportunities that exist when investment experts over and under-react to certain news stories. By analyzing breaking financial news articles and focusing on specific parts of speech, portfolio selection, term weighting and even article sentiment, the AZFinText system becomes a powerful tool and is a radically different way of looking at stock market prediction.
This system differs from other systems in that it uses financial text as one of its key means of predicting stock price movement. This reduces the information lag-time problem evident in many similar systems where new information must be transcribed (e.g., such as losing a costly court battle or having a product recall), before the quant can react appropriately. AZFinText overcomes these limitations by utilizing the terms used in financial news articles to predict future stock prices twenty minutes after the news article has been released.<ref>Schumaker, R., (2006). [http://www.robschumaker.com/publications/AMCIS%20-%20Textual%20Analysis%20of%20Stock%20Market%20Prediction%20Using%20Financial%20News%20Articles.pdf Textual Analysis of Stock Market Prediction Using Financial News Articles]. ''12th Americas Conference on Information Systems (AMCIS-2006)'', Aug 2006. Acapulco, Mexico.</ref>


It is believed that certain article terms can move stocks more than others. Terms such as ''factory exploded'' or ''workers strike'' will have a depressing effect on stock prices whereas terms such as ''earnings rose'' will tend to increase stock prices.
⚫
== Overview of Research ==
⚫
The foundation of AZFinText can be found in the ACM TOIS article. <ref>Schumaker, R. and Chen H., (2009). [http://portal.acm.org/citation.cfm?id=1462198.1462204 Textual Analysis of Stock Market Prediction Using Breaking Financial News]. ''Association for Computing Machinery Transactions on Information Systems'', 27(2).</ref> Within this paper, the authors tested several different prediction models and linguistic textual representations. From this work, it was found that using the article terms and the price of the stock at the time the article was released was the most effective model and using proper nouns was the most effective textual representation technique. Combining the two, AZFinText netted a 2.84% trading return over the five-week study period.


When a human trading expert sees certain terms, they will react in a somewhat predictable fashion. AZFinText capitalizes on the arbitrage opportunities that exist when investment experts over and under-react to certain news stories. By analyzing breaking financial news articles and focusing on specific parts of speech, portfolio selection, term weighting and even article sentiment, the AZFinText system becomes a powerful tool and is a radically different way of looking at stock market prediction.
⚫
AZFinText was then extended to study what combination of peer organizations help to best train the system. <ref>Schumaker, R. and Chen, H., (2009). [http://portal.acm.org/citation.cfm?id=1570564 A Quantitative Stock Prediction System based on Financial News]. ''Information Processing and Management'', 45(5): 571-583.</ref> Using the premise that [[IBM]] has more in common with [[Microsoft]] than [[GM]], AZFinText studied the effect of varying peer-based training sets. To do this, AZFinText trained on the various levels of [[GICS]] and evaluated the results. It was found that sector-based training was most effective, netting an 8.50% trading return, outperforming [[Jim Cramer]], Jim Jubak and DayTraders.com during the study period. AZFinText was also compared against the top 10 quantitative systems and outperformed 6 of them.


⚫
== Overview of research ==
⚫
A third study investigated the role of portfolio building in a textual financial prediction system. <ref>Schumaker, R. and Chen, H., (2008). [http://portal.acm.org/citation.cfm?id=1331274 Evaluating a News-Aware Quantitative Trader: The Effects of Momentum and Contrarian Stock Selection Strategies]. ''Journal of the American Society for Information Science and Technology'', 59(2): 247-255.</ref> From this study, Momentum and Contrarian stock portfolios were created and tested. Using the premise that past winning stocks will continue to win and past losing stocks will continue to lose, AZFinText netted a 20.79% return during the study period. It was also noted that traders were generally over-reacting to news events, creating the opportunity of abnormal returns.
⚫
The foundation of AZFinText can be found in the ACM TOIS article.<ref>Schumaker, R. and Chen H., (2009). [http://portal.acm.org/citation.cfm?id=1462198.1462204 Textual Analysis of Stock Market Prediction Using Breaking Financial News]. ''Association for Computing Machinery Transactions on Information Systems'', 27(2).</ref> Within this paper, the authors tested several different prediction models and linguistic textual representations. From this work, it was found that using the article terms and the price of the stock at the time the article was released was the most effective model and using proper nouns was the most effective textual representation technique. Combining the two, AZFinText netted a 2.84% trading return over the five-week study period.


⚫
AZFinText was then extended to study what combination of peer organizations help to best train the system.<ref>Schumaker, R. and Chen, H., (2009). [http://portal.acm.org/citation.cfm?id=1570564 A Quantitative Stock Prediction System based on Financial News]. ''Information Processing and Management'', 45(5): 571-583.</ref> Using the premise that [[IBM]] has more in common with [[Microsoft]] than [[General Motors|GM]], AZFinText studied the effect of varying peer-based training sets. To do this, AZFinText trained on the various levels of [[Global Industry Classification Standard|GICS]] and evaluated the results. It was found that sector-based training was most effective, netting an 8.50% trading return, outperforming [[Jim Cramer]], Jim Jubak and DayTraders.com during the study period. AZFinText was also compared against the top 10 quantitative systems and outperformed 6 of them.
⚫
A fourth study looked into using author sentiment as an added predictive variable. <ref>Schumaker, R., Zhang, Y. and Huang, C., (2008). [http://www.robschumaker.com/publications/IIMA%20-%20Sentiment%20Analysis%20of%20Financial%20News%20Articles.pdf Sentiment Analysis of Financial News Articles]. ''20th Annual Conference of International Information Management Association'', Oct. 2009. Houston, TX.</ref> Using the premise that an author can unwittingly influence market trades simply by the terms they use, AZFinText was tested using tone and polarity features. It was found that Contrarian activity was occurring within the market, where articles of a positive tone would decrease in price and articles of a negative tone would increase in price.


⚫
A third study investigated the role of portfolio building in a textual financial prediction system.<ref>Schumaker, R. and Chen, H., (2008). [http://portal.acm.org/citation.cfm?id=1331274 Evaluating a News-Aware Quantitative Trader: The Effects of Momentum and Contrarian Stock Selection Strategies]. ''Journal of the American Society for Information Science and Technology'', 59(2): 247-255.</ref> From this study, Momentum and Contrarian stock portfolios were created and tested. Using the premise that past winning stocks will continue to win and past losing stocks will continue to lose, AZFinText netted a 20.79% return during the study period. It was also noted that traders were generally overreacting to news events, creating the opportunity of abnormal returns.
⚫
A further study investigated what article verbs have the most influence on stock price movement. <ref>Schumaker, R., (2010). [http://www.robschumaker.com/publications/SM%20-%20An%20Analysis%20of%20Verbs%20in%20Financial%20News%20Articles%20and%20their%20Impact%20on%20Stock%20Price.pdf An Analysis of Verbs in Financial News Articles and their Impact on Stock Price]. ''NAACL Workshop on Social Media and Computational Linguistics'', June 2010. Los Angeles, CA.</ref> From this work, it was found that ''planted'', ''announcing'', ''front'', ''smaller'' and ''crude'' had the highest positive impact on stock price.


⚫
A fourth study looked into using author sentiment as an added predictive variable.<ref>Schumaker, R., Zhang, Y. and Huang, C., (2008). [http://www.robschumaker.com/publications/IIMA%20-%20Sentiment%20Analysis%20of%20Financial%20News%20Articles.pdf Sentiment Analysis of Financial News Articles]. ''20th Annual Conference of International Information Management Association'', Oct. 2009. Houston, TX.</ref> Using the premise that an author can unwittingly influence market trades simply by the terms they use, AZFinText was tested using tone and polarity features. It was found that Contrarian activity was occurring within the market, where articles of a positive tone would decrease in price and articles of a negative tone would increase in price.
⚫
== Notable Publicity ==

Because of its revolutionary way of looking at price prediction, AZFinText has been the topic of discussion by numerous media outlets in many different languages. Some of the more notable ones include [[Slashdot]], [[Technology_Review|MIT's Technology Review]], [[Motley Fool]], Crossing Wall Street, [[WBIR]] in Knoxville, TN and Motherboard TV.
⚫
A further study investigated what article verbs have the most influence on stock price movement.<ref>Schumaker, R., (2010). [http://www.robschumaker.com/publications/SM%20-%20An%20Analysis%20of%20Verbs%20in%20Financial%20News%20Articles%20and%20their%20Impact%20on%20Stock%20Price.pdf An Analysis of Verbs in Financial News Articles and their Impact on Stock Price]. ''NAACL Workshop on Social Media and Computational Linguistics'', June 2010. Los Angeles, CA.</ref> From this work, it was found that ''planted'', ''announcing'', ''front'', ''smaller'' and ''crude'' had the highest positive impact on stock price.

⚫
== Notable publicity ==
AZFinText has been the topic of discussion by numerous media outlets. Some of the more notable ones include ''[[The Wall Street Journal]]'',<ref>{{Cite web |url=https://blogs.wsj.com/digits/2010/06/21/using-artificial-intelligence-to-digest-financial-news/ |title=Using Artificial Intelligence to Digest News, Trade Stocks |last=Valentino-DeVries |first=Jennifer |website=The Wall Street Journal |date=June 21, 2010 |access-date=January 20, 2017}}</ref> ''[[Technology Review|MIT's Technology Review]]'',<ref name="MIT Technology Review">{{cite journal |url=https://www.technologyreview.com/2010/06/10/202923/ai-that-picks-stocks-better-than-the-pros/ |last=Mims |first=C. |title=AI That Picks Stocks Better Than the Pros |journal=MIT Technology Review |date=June 10, 2010}}</ref> [[Dow Jones Newswire]],<ref name="Dow Jones Newswire">{{cite journal |last=Subsin |first=N. |title=Algorithmic and Trading Products Newsletter |journal=Dow Jones Newswire |date=Nov 24, 2010}}</ref> [[WBIR]]<ref name="WBIR">{{cite journal |title=AI that Picks Stocks Better than the Pros |journal=WBIR |date=June 10, 2010}}</ref> in Knoxville, TN, [[Slashdot]]<ref name="Slashdot">{{cite journal |title=Quant AI Picks Stocks Better than Humans |journal=Slashdot |date=June 12, 2010}}</ref> and other media outlets.


== References ==
== References ==
{{reflist}}
<!--- See http://en.wikipedia.org/wiki/Wikipedia:Footnotes on how to create references using <ref></ref> tags which will then appear here automatically -->
{{Reflist}}


== External links ==
== External links ==
* [https://blogs.wsj.com/digits/2010/06/21/using-artificial-intelligence-to-digest-financial-news/ https://blogs.wsj.com/digits/2010/06/21/using-artificial-intelligence-to-digest-financial-news/]
* [http://slashdot.org/story/10/06/12/1341212/Quant-AI-Picks-Stocks-Better-Than-Humans slashdot.org/story/10/06/12/1341212/Quant-AI-Picks-Stocks-Better-Than-Humans]
* [http://slashdot.org/story/10/06/12/1341212/Quant-AI-Picks-Stocks-Better-Than-Humans slashdot.org/story/10/06/12/1341212/Quant-AI-Picks-Stocks-Better-Than-Humans]
* [http://www.technologyreview.com/blog/guest/25308/ www.technologyreview.com/blog/guest/25308/]
* [http://www.technologyreview.com/blog/guest/25308/ www.technologyreview.com/blog/guest/25308/]


[[Category:Mathematical finance]]

[[Category:Artificial intelligence]]
<!--- Categories --->
[[Category:Articles created via the Article Wizard]]
[[Category:Artificial intelligence engineering]]

Latest revision as of 14:28, 18 November 2024

Arizona Financial Text System (AZFinText) is a textual-based quantitative financial prediction system written by Robert P. Schumaker of University of Texas at Tyler and Hsinchun Chen of the University of Arizona.

System

[edit]

This system differs from other systems in that it uses financial text as one of its key means of predicting stock price movement. This reduces the information lag-time problem evident in many similar systems where new information must be transcribed (e.g., such as losing a costly court battle or having a product recall), before the quant can react appropriately. AZFinText overcomes these limitations by utilizing the terms used in financial news articles to predict future stock prices twenty minutes after the news article has been released.[1]

It is believed that certain article terms can move stocks more than others. Terms such as factory exploded or workers strike will have a depressing effect on stock prices whereas terms such as earnings rose will tend to increase stock prices.

When a human trading expert sees certain terms, they will react in a somewhat predictable fashion. AZFinText capitalizes on the arbitrage opportunities that exist when investment experts over and under-react to certain news stories. By analyzing breaking financial news articles and focusing on specific parts of speech, portfolio selection, term weighting and even article sentiment, the AZFinText system becomes a powerful tool and is a radically different way of looking at stock market prediction.

Overview of research

[edit]

The foundation of AZFinText can be found in the ACM TOIS article.[2] Within this paper, the authors tested several different prediction models and linguistic textual representations. From this work, it was found that using the article terms and the price of the stock at the time the article was released was the most effective model and using proper nouns was the most effective textual representation technique. Combining the two, AZFinText netted a 2.84% trading return over the five-week study period.

AZFinText was then extended to study what combination of peer organizations help to best train the system.[3] Using the premise that IBM has more in common with Microsoft than GM, AZFinText studied the effect of varying peer-based training sets. To do this, AZFinText trained on the various levels of GICS and evaluated the results. It was found that sector-based training was most effective, netting an 8.50% trading return, outperforming Jim Cramer, Jim Jubak and DayTraders.com during the study period. AZFinText was also compared against the top 10 quantitative systems and outperformed 6 of them.

A third study investigated the role of portfolio building in a textual financial prediction system.[4] From this study, Momentum and Contrarian stock portfolios were created and tested. Using the premise that past winning stocks will continue to win and past losing stocks will continue to lose, AZFinText netted a 20.79% return during the study period. It was also noted that traders were generally overreacting to news events, creating the opportunity of abnormal returns.

A fourth study looked into using author sentiment as an added predictive variable.[5] Using the premise that an author can unwittingly influence market trades simply by the terms they use, AZFinText was tested using tone and polarity features. It was found that Contrarian activity was occurring within the market, where articles of a positive tone would decrease in price and articles of a negative tone would increase in price.

A further study investigated what article verbs have the most influence on stock price movement.[6] From this work, it was found that planted, announcing, front, smaller and crude had the highest positive impact on stock price.

Notable publicity

[edit]

AZFinText has been the topic of discussion by numerous media outlets. Some of the more notable ones include The Wall Street Journal,[7] MIT's Technology Review,[8] Dow Jones Newswire,[9] WBIR[10] in Knoxville, TN, Slashdot[11] and other media outlets.

References

[edit]
  1. ^ Schumaker, R., (2006). Textual Analysis of Stock Market Prediction Using Financial News Articles. 12th Americas Conference on Information Systems (AMCIS-2006), Aug 2006. Acapulco, Mexico.
  2. ^ Schumaker, R. and Chen H., (2009). Textual Analysis of Stock Market Prediction Using Breaking Financial News. Association for Computing Machinery Transactions on Information Systems, 27(2).
  3. ^ Schumaker, R. and Chen, H., (2009). A Quantitative Stock Prediction System based on Financial News. Information Processing and Management, 45(5): 571-583.
  4. ^ Schumaker, R. and Chen, H., (2008). Evaluating a News-Aware Quantitative Trader: The Effects of Momentum and Contrarian Stock Selection Strategies. Journal of the American Society for Information Science and Technology, 59(2): 247-255.
  5. ^ Schumaker, R., Zhang, Y. and Huang, C., (2008). Sentiment Analysis of Financial News Articles. 20th Annual Conference of International Information Management Association, Oct. 2009. Houston, TX.
  6. ^ Schumaker, R., (2010). An Analysis of Verbs in Financial News Articles and their Impact on Stock Price. NAACL Workshop on Social Media and Computational Linguistics, June 2010. Los Angeles, CA.
  7. ^ Valentino-DeVries, Jennifer (June 21, 2010). "Using Artificial Intelligence to Digest News, Trade Stocks". The Wall Street Journal. Retrieved January 20, 2017.
  8. ^ Mims, C. (June 10, 2010). "AI That Picks Stocks Better Than the Pros". MIT Technology Review.
  9. ^ Subsin, N. (Nov 24, 2010). "Algorithmic and Trading Products Newsletter". Dow Jones Newswire.
  10. ^ "AI that Picks Stocks Better than the Pros". WBIR. June 10, 2010.
  11. ^ "Quant AI Picks Stocks Better than Humans". Slashdot. June 12, 2010.
[edit]
Retrieved from "https://en.wikipedia.org/enwiki/w/index.php?title=AZFinText&oldid=1258170877"