DOCSLIB.ORG

Stock Portfolio Structure of Individual Investors Infers Future Trading

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Stock Portfolio Structure of Individual Investors Infers Future Trading Stock portfolio structure of individual investors infers future trading behavior Ludvig Bohlin∗ and Martin Rosvall† Integrated Science Lab, Ume˚a University, Sweden (Dated: July 28, 2014) Although the understanding of and motivation behind individual trading behavior is an important puzzle in finance, little is known about the connection between an investor’s portfolio structure and her trading behavior in practice. In this paper, we investigate the relation between what stocks investors hold, and what stocks they buy, and show that investors with similar portfolio structures to a great extent trade in a similar way. With data from the central register of shareholdings in Sweden, we model the market in a similarity network, by considering investors as nodes, connected with links representing portfolio similarity. From the network, we find investor groups that not only identify different investment strategies, but also represent individual investors trading in a similar way. These findings suggest that the stock portfolios of investors hold meaningful information, which could be used to earn a better understanding of stock market dynamics. Introduction explained by homophily, i.e., the tendency of individuals Stock market trading provides opportunities at the cost to behave and bond with others who are similar. In- of risk. For investors, the ultimate trading goal is to make vestors potentially trade more similarly if they share cer- as much money as possible by acting in such a way that tain properties, including, for example, age (12), gender the highest possible profit is realized at minimal risk. Yet (13) and familiarity (14). However, classifying individ- how to trade optimally is far from obvious, and many fac- ual investors into such distinct categories is not straight- tors influence trading behavior. The traditional approach forward. The classification is in some cases motivated to represent market trading has been a model perspec- by theoretical considerations, and in other cases by ob- tive, assuming that investors act as rational and identical served patterns in the data. Categories include, for ex- agents (1, 2). However, this assumption has been chal- ample, fundamentalists and chartists (15), and informed lenged by empirical evidence, which suggests that also and uninformed traders (16). Other examples of investor other elements are present in financial markets (3). For categorizations in financial data are derived from trad- example, researchers have suggested models in which the ing correlations (17), direct stock trading data (18), and economic decisions of investors also consider the effects network approaches (19, 20). of social, cognitive, and emotional factors. These factors Although the understanding of and motivation behind and their influence on trading are often studied with data individual trading behavior is an important puzzle in fi- external to the actual trading process and include, for ex- nance, little is known about the connection between in- ample, proximity (4), social media interactions (5) and vestors’ portfolio structure and their trading behavior in web engine search queries (6, 7), In this work, we instead practice. Studies have found that many individual in- focus on data more directly relevant to the trading by vestors tend to hold poorly diversified portfolios and in- relating trading patterns and behavioral biases to stock stead concentrate investments in only a few stocks (21). portfolio structure. With real financial data on individ- The difficulty of searching through all available stocks ual investors from the Swedish stock market, we study also makes it more likely for individual investors to in- the connection between what stocks investors hold, and vest in stocks that attract their attention (22), and these arXiv:1402.2494v2 [q-fin.GN] 28 Aug 2014 what stocks they buy. attention-grabbing stocks are typically the ones that in- Trading by individual investors and related behavioral vestors already hold in their portfolios. This bias indi- biases have been studied by, for example, local bias (8), cates that portfolio structure and trading decisions are overconfidence (9), sensation seeking (10) and the dis- naturally connected. position effect (11). Studies show that investors base In this paper, we explore the connection between port- trading not only on rationale information, but are also folio structure and trading behavior in individual in- affected by factors connected to both personal character- vestors. With detailed data on stock portfolios from the istics and associated external conditions. Various indi- central register of shareholdings in Sweden, we study the vidual biases give rise to a trading heterogeneity among relation between stock portfolio similarity and trading investors, or among groups of investors. The presence of similarity. Unlike most previous research, we do not an- such investor groups with similar trading behavior can be alyze the direct trading, but instead focus on long-term trading behavior by looking at changes in share portfo- lios over time. We aim to examine two main questions: (1) How do investors in the stock market structure their ∗[email protected]; Corresponding author portfolios? And (2) Can we learn about trading behavior †[email protected] by looking at the investors’ portfolio structure? 2 To answer the questions of individual trading behavior holdings. Direct holdings are registered in the investor’s we take three steps: First, we investigate how individ- name, as opposed to nominee holdings, which are reg- ual investors hold stocks, and how they trade. Second, istered and managed by an equity manager on behalf we divide investors into groups based on portfolio sim- of the investor. The direct holdings of all investors are ilarity. This division is done with a network approach, presented in each half-year report, with detailed infor- where individual investors are considered to be nodes, mation about registration type, share amount, and the and links between investors are constructed according to equity ISIN in which the shares are held. This infor- stock portfolio similarities. To group similar investors, mation makes it possible to find share changes in the we analyze the network with the community detection portfolios between reports, provided that investors have algorithm Infomap (23). Third, we analyze the derived a traceable identification number. Investors who lack a groups to investigate the relationship between portfolio Swedish identification can not be reliably tracked in the structure and trading behavior. This analysis is done data over time, and we therefore excluded these investors by comparing investor trading within groups to investor in the analysis. trading outside the group. In the following section we To reduce the effects of noise in the data, some con- present the methods and the results, and, in short, we ditions for the included stocks were established. First, find that the portfolio structure of individual investors we only considered stocks of companies that existed for holds information on trading behavior, and that investors the entire time period. We therefore excluded stocks that with similar portfolios, to a great extent, trade in a sim- were introduced or removed from the market during the ilar way. time period for any reason. This exclusion was done to enable comparisons between two share reports without Methods changes in the company domain. Furthermore, we also Data from the central register of Swedish required that the total share amount of a stock must not shareholdings have changed more than five percent during the time pe- We examined more than 100,000 individual investors who riod. This condition was set because larger changes make were actively trading in the Swedish stock market from it hard to distinguish actual active trades of investors 2009 to 2011. The investors and their stock portfolios from more passive changes in the portfolios directly re- were extracted from a dataset with around two million lated to a share amount change, as, for example, in the investors. The dataset stems from the central register case of stock splitting. As a consequence of the share of shareholdings in Sweden, and covers basically all in- amount change criteria, we excluded, for example, the vestors and their holdings in every publicly traded com- companies H&M and Swedish Match from the analysis. pany in Sweden. The dataset was provided by Euro- Finally, only listed stocks were considered in the analysis, clear Sweden AB, and permission to use the data was since these stocks are publicly traded and it is possible to given under a special agreement. Data are presented in find an explicit price for them. It is also worth noticing half-year share register reports between June 30, 2009, the distinction between stocks designated A and B on and December 30, 2011, with detailed ownership infor- the Swedish market. A company can be associated with mation of investors in each registered company. The re- more than one stock, because A and B stocks, and other ports also included the companies’ total share amount potential stock classes in a company, must have different and their corresponding stock ISIN (International Secu- ISIN codes. We considered these different stock classes rities Identification Number). Additional data, obtained as separate stocks, because classes with less voting power from the Swedish Central Statistics Office (SCB), pro- usually are more liquid, and therefore give rise to differ- vided share prices for companies listed on the Stockholm ent trading than the ones with superior voting rights. stock exchange. Those data specify share prices at stock In summary, we examined investors on the Swedish exchange closing time, i.e., the price of the latest sold stock market who are natural persons, traceable over share on the last trading day. If price data are lacking, time, active in trading and primarily registered as share- bid price and then ask price were used instead. In total, holders. This selection means that we, for example, ex- the data contain share prices for around 500 listed stocks.
Load more

Recommended publications
  • Program, List of Abstracts, and List of Participants and of Participating Institutions
    Wellcome to the Econophysics Colloquium 2016! In this booklet you will find relevant information about the colloquium, like the program, list of abstracts, and list of participants and of participating institutions. We hope you all have a pleasant and productive time here in São Paulo! Program Wednesday, 27 Thursday, 28 Friday, 29 8.30 – Registration 9:15 am 9:15 – Welcome 9.30 am Universality in the interoccurrence Multiplex dependence structure of Measuring economic behavior using 9.30 - times in finance and elsewhere financial markets online data 10.30 am (C. Tsallis) (T. Di Matteo) (S. Moat) 10.30- Coffee Break Coffee Break Coffee Break 11.00 am Portfolio optimization under expected Financial markets, self-organized Sensing human activity using online 11.00- shortfall: contour maps of estimation criticality and random strategies data 12.00 am error (A. Rapisarda) (T. Preis) (F. Caccioli) 12.00- Lunch Lunch Lunch 2.00 pm Trading networks at NASDAQ OMX Complexity driven collapses in large 2.00-3.00 IFT-Colloquium Helsinki random economies pm (R. Mantegna) (R. Mantegna) (G. Livan) Financial market crashes can be 3.00-4.00 Poster Session quantitatively forecasted Parallel Sessions 2A and 2B pm (S.A. Cheong) 4.00-4.30 Coffee Break Coffee Break Closing pm Macroeconomic modelling with Discussion Group: Financial crises and 4.30-5.30 heterogeneous agents: the master systemic risk - Presentation by Thiago pm equation approach (M. Grasselli) Christiano Silva (Banco Central) 5.45-6.45 Discussion Group: Critical transitions in Parallel Sessions 1A and 1B pm markets 7.00- Dinner 10.00 pm Wednesday, 27 July Plenary Sessions (morning) Auditorium 9.30 am – 12 pm 9.30-10.30 am Universality in the Interoccurence times in finance and elsewhere Constantino Tsallis (CBPF, Brazil) A plethora of natural, artificial and social systems exist which do not belong to the Boltzmann-Gibbs (BG) statistical-mechanical world, based on the standard additive entropy SBG and its associated exponential BG factor.
    [Show full text]
  • WRAP 0265813516687302.Pdf
    Original citation: Seresinhe, Chanuki Illushka, Moat, Helen Susannah and Preis, Tobias. (2017) Quantifying scenic areas using crowdsourced data. Environment and Planning B : Urban Analytics and City Science . 10.1177/0265813516687302 Permanent WRAP URL: http://wrap.warwick.ac.uk/87375 Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes this work of researchers of the University of Warwick available open access under the following conditions. This article is made available under the Creative Commons Attribution 3.0 (CC BY 3.0) license and may be reused according to the conditions of the license. For more details see: http://creativecommons.org/licenses/by/3.0/ A note on versions: The version presented in WRAP is the published version, or, version of record, and may be cited as it appears here. For more information, please contact the WRAP Team at: [email protected] warwick.ac.uk/lib-publications Urban Analytics and Article City Science Environment and Planning B: Urban Quantifying scenic areas Analytics and City Science 0(0) 1–16 ! The Author(s) 2017 using crowdsourced data Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0265813516687302 Chanuki Illushka Seresinhe, journals.sagepub.com/home/epb Helen Susannah Moat and Tobias Preis Warwick Business School, University of Warwick, UK; The Alan Turing Institute, UK Abstract For centuries, philosophers, policy-makers and urban planners have debated whether aesthetically pleasing surroundings can improve our wellbeing. To date, quantifying how scenic an area is has proved challenging, due to the difficulty of gathering large-scale measurements of scenicness. In this study we ask whether images uploaded to the website Flickr, combined with crowdsourced geographic data from OpenStreetMap, can help us estimate how scenic people consider an area to be.
    [Show full text]
  • Training Tomorrow's Analysts
    64 www.efmd.org/globalfocus 10101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101 01010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010 10101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101 01010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010 10101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101 01010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010 10101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101010101
    [Show full text]
  • The Digital Treasure Trove
    background in computer science. My PhD gave me the a large scale,” says Preis. “Google provides the possibility chance to bring all these different disciplines together, of looking at the early stages of collective decision making. to try and understand the vast amounts of data generated Investors are not disconnected from the world, they’re by the financial world. So everything I’m doing today is hugely Googling and using various platforms and services to collect interdisciplinary, involving large volumes of data.” information for their decisions.” In effect, these new sources In the era of big data, Preis’s research looks at what of data can predict human behaviour. The events of the world he calls the “digital traces”, the digital detritus, the granules are Googled before they happen. The next challenge for these of information and insight we leave behind in our interactions researchers is to be able to identify key words and phrases at with technology and the internet in particular. His research the time they are being used rather than historically. is probably best described as “computational social science”, an emerging interdisciplinary field which aims to exploit such The risks of diversification data in order to better understand how our complex social world operates. Preis’s fascination with big data and financial markets also led to his involvement in a study of the US market index, the Dow Positive traces Jones. His research analysed the daily closing prices of the 30 stocks forming the Dow Jones Industrial Average from March One intriguing source of information used by Preis’s research 15, 1939, to December 31, 2010.
    [Show full text]
  • Quantifying Trading Behavior in Financial Markets Using Google Trends : Scientific Reports : Nature Publishing Group
    13-4-27 Quantifying Trading Behavior in Financial Markets Using Google Trends : Scientific Reports : Nature Publishing Group nature.com Sitemap Cart Login Register Search 2013 April Article SCIENTIFIC REPORTS | ARTICLE OPEN Quantifying Trading Behavior in Financial Markets Using Google Trends Tobias Preis, Helen Susannah Moat & H. Eugene Stanley Scientific Reports 3, Article number: 1684 doi:10.1038/srep01684 Received 25 February 2013 Accepted 03 April 2013 Published 25 April 2013 Crises in financial markets affect humans worldwide. Detailed market data on trading decisions reflect some of thPerint complex human behavior that has led to these crises. We suggest that massive new data sources resulting from human interaction with the Internet may offer a new perspective on the behavior of market participants in periods of large market movements. By analyzing changes in Google query volumes for search terms related to finance, we find patterns that may be interpreted as “early warning signs” of stock market moves. Our results illustrate the potential that combining extensive behavioral data sets offers for a better understanding of collective human behavior. Subject terms: Computational science Applied physics Information theory and computation Statistical physics, thermodynamics and nonlinear dynamics Introduction The increasing volumes of ‘big data’ reflecting various aspects of our everyday activities represent a vital new opportunity for scientists to address fundamental questions about the complex world we inhabit1, 2, 3, 4, 5, 6, 7. Financial markets are a prime target for such quantitative investigations8, 9. Movements in the markets exert immense impacts on personal fortunes and geopolitical events, generating considerable scientific attention to this subject10, 11, 12, 13, 14, 15, 16, 17, 18, 19.
    [Show full text]
  • Analysis of Financial News with Newsstream
    Date: 2015-11-01 IJS-DP-11965 Analysis of Financial News with NewsStream Petra Kralj Novak, Miha Grˇcar,Borut Sluban and Igor Mozetiˇc JoˇzefStefan Institute Ljubljana, Slovenia petra.kralj.novak, miha.grcar, borut.sluban, igor.mozetic @ijs.si { } Abstract Unstructured data, such as news and blogs, can provide valuable in- sights into the financial world. We present the NewsStream portal, an in- tuitive and easy-to-use tool for news analytics, which supports interactive querying and visualizations of the documents at different levels of detail. It relies on a scalable architecture for real-time processing of a continu- ous stream of textual data, which incorporates data acquisition, cleaning, natural-language preprocessing and semantic annotation components. It has been running for over two years and collected over 18 million news articles and blog posts. The NewsStream portal can be used to answer the questions when, how often, in what context, and with what sentiment was a financial entity or term mentioned in a continuous stream of news and blogs, and therefore providing a complement to news aggregators. We illustrate some features of our system in four use cases: relations between the rating agencies and the PIIGS countries, reflection of financial news on credit default swap (CDS) prices, the emergence of the Bitcoin digital currency, and visualizing how the world is connected through news. Keywords: stream processing architecture, unstructured data, financial news, sentiment analysis, ontology-based entity recognition, visual analytics 1 Introduction Tens of thousands of news articles are being published daily on the world wide arXiv:1508.00027v2 [cs.IR] 7 Nov 2015 web by thousands of news sources.
    [Show full text]
  • Estimating Tourism Statistics with Wikipedia Page Views
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Warwick Research Archives Portal Repository Original citation: Alis, Christian M., Letchford, Adrian, Moat, Helen Susannah and Preis, Tobias (2015) Estimating tourism statistics with Wikipedia page views. In: WebSci '15 Web Science Conference, Oxford, 28 Jun - 1 Jul 2015. Published in: WebSci '15 Proceedings of the ACM Web Science Conference pp. Article 33. Permanent WRAP URL: http://wrap.warwick.ac.uk/79613 Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes this work by researchers of the University of Warwick available open access under the following conditions. Copyright © and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable the material made available in WRAP has been checked for eligibility before being made available. Copies of full items can be used for personal research or study, educational, or not-for profit purposes without prior permission or charge. Provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way. Publisher’s statement: "© ACM, 2015. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in WebSci '15 Proceedings of the ACM Web Science Conference http://doi.acm.org/10.1145/2786451.2786925 A note on versions: The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher’s version.
    [Show full text]
  • Predicting the Present with Google Trends
    Predicting the Present with Google Trends Hyunyoung Choi, Hal Varian∗ December 18, 2011 Abstract In this paper we show how to use search engine data to forecast near-term values of economic indicators. Examples include automobile sales, unemployment claims, travel destination planning, and consumer confidence. Government agencies periodically release indicators of the level of economic activity in various sectors. However, these releases are typically only available with a reporting lag of several weeks and are often revised a few months later. It would clearly be helpful to have more timely forecasts of these economic indicators. Nowadays there are several sources of data on real-time economic activity available from private sector companies such as Google, MasterCard, Federal Express, UPS, Intuit and many others. In this paper we examine Google Trends, which is a a real-time daily and weekly index of the volume of queries that users enter into Google. We have found that these query indices are often correlated with various economic indicators and may be helpful for short-term economic prediction. We are not claiming that Google Trends data can help in predicting the future. Rather we are claiming that Google Trends may help in predicting the present. For example, the volume of queries on automobile sales during the second week in June may be helpful in predicting the June auto sales report which is released several weeks later in July. It may also be true that June queries help to predict July sales, but we leave that question for future research, as this depends very much on the particular time series in question.
    [Show full text]
  • Using News Articles to Predict Stock Price Returns
    Using News Articles to Predict Stock Price Returns Ricardo Herrmann Rodrigo Togneri [email protected] [email protected] Luciano Tozato Wei Lin [email protected] [email protected] Knowledge Sharing Article © 2017 Dell Inc. or its subsidiaries. Table of Contents Overview ....................................................................................................................... 3 Introduction ................................................................................................................... 4 Related Work ............................................................................................................. 6 Deep CNN model for text classification ......................................................................... 6 Empirical evaluation ...................................................................................................... 9 Data sources ........................................................................................................... 10 Stock prices ......................................................................................................... 10 News articles ........................................................................................................ 10 Pre-trained word embeddings .............................................................................. 11 Results .................................................................................................................... 12 Conclusion .................................................................................................................
    [Show full text]
  • Quantifying Economic Behaviour Using Big Data
    Quantifying Economic Behaviour Using Big Data Tobias Preis Warwick Business School [email protected] www.tobiaspreis.de A map of the world built only from GPS locations of Flickr photos The big data explosion 1 The advantage of looking forward Suzy Moat & Tobias Preis Future Orientation Index 2010 Based on Preis, Moat, Stanley and Bishop (2012) more Google searches for “2009” more Google searches for “2011” Suzy Moat & Tobias Preis Future Orientation Index 2012 Based on Preis, Moat, Stanley and Bishop (2012) 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Ratio of Google searches for “2011” to searches for “2009” during 2010 for 45 countries 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 FutureRatio Orientation of Google searches for “2013” to searches Index for “2011” during 20122010 for 45 countries Richer countries look forward A B Preis, Moat, Stanley & 5 “2007” “2008” “2009” 4 Bishop (2012) USD] 0 4 10 3 2 5 “2009” “2010” “2011” Featured by: 1 Search Volume 0 2008 2009 2010 GDP / Capita [10 0.0 0.5 1.0 1.5 2.0 Time with Weekly Granularity Future-Orientation Index Photo: Perpetual Tourist 2 Predicting stock markets Financial markets: big data Photo: Perpetual Tourist The Internet: big data Moat et al. (2013); Hypothetical strategy Preis et al. (2013) number of Google searches for keyword week t Moat et al. (2013); Hypothetical strategy Preis et al. (2013) number of Google searches for keyword week t-3 t-2 t-1 t Moat et al.
    [Show full text]
  • Using Machine Learning and Alternative Data to Predict Movements in Market Risk
    USING MACHINE LEARNING AND ALTERNATIVE DATA TO PREDICT MOVEMENTS IN MARKET RISK Thomas Dierckx Jesse Davis Department of Statistics Department of Computer Science Department of Computer Science KU Leuven, Belgium KU Leuven, Belgium [email protected] [email protected] Wim Schoutens Department of Statistics KU Leuven, Belgium [email protected] September 18, 2020 ABSTRACT Using machine learning and alternative data for the prediction of financial markets has been a pop- ular topic in recent years. Many financial variables such as stock price, historical volatility and trade volume have already been through extensive investigation. Remarkably, we found no existing research on the prediction of an asset’s market implied volatility within this context. This forward- looking measure gauges the sentiment on the future volatility of an asset, and is deemed one of the most important parameters in the world of derivatives. The ability to predict this statistic may therefore provide a competitive edge to practitioners of market making and asset management alike. Consequently, in this paper we investigate Google News statistics and Wikipedia site traffic as al- ternative data sources to quantitative market data and consider Logistic Regression, Support Vector Machines and AdaBoost as machine learning models. We show that movements in market implied volatility can indeed be predicted through the help of machine learning techniques. Although the em- ployed alternative data appears to not enhance predictive accuracy, we reveal preliminary evidence of non-linear relationships between features obtained from Wikipedia page traffic and movements in market implied volatility. Keywords Implied Volatility · Machine Learning · Alternative Data arXiv:2009.07947v1 [q-fin.CP] 16 Sep 2020 1 Introduction To predict the stock market, statisticians generally relied on econometric methods.
    [Show full text]
  • Google Searches Hold Key to Future Market Crashes 28 July 2014
    Google searches hold key to future market crashes 28 July 2014 A team of researchers from Warwick Business identify patterns in search activity that might be School and Boston University have developed a related to subsequent real world behaviour, the method to automatically identify topics that people team quantified the meaning of every single word search for on Google before subsequent stock on Wikipedia. This allowed the researchers to market falls. categorize words into topics, so that a "business" topic may contain words such as "business", Applied to data between 2004 and 2012, the "management", and "bank". The algorithm identified method shows that increases in searches for a broad selection of topics, ranging from food to business and politics preceded falls in the stock architecture to cricket. market. The study, 'Quantifying the semantics of search behavior before stock market moves,' was The team then used Google Trends to see how published in the Proceedings of the National often each week thousands of these words were Academy of Sciences. searched for by Internet users in the United States between 2004 and 2012. By using these search The researchers suggest that this method could be activity datasets in a simple trading strategy for the applied to help identify warning signs in search S&P 500, they found that changes in how often data before a range of real world events. users searched for terms relating to business and politics could be connected to subsequent stock "Search engines, such as Google, record almost market moves. everything we search for," said Chester Curme, Research Fellow at Warwick Business School and "By mining these datasets, we were able to identify lead author of the study.
    [Show full text]