Elsevier

Decision Support Systems

Volume 59, March 2014, Pages 52-62
Decision Support Systems

A decision support system for stock investment recommendations using collective wisdom

https://doi.org/10.1016/j.dss.2013.10.005Get rights and content

Highlights

  • Stock ratings of online stock communities have shown to provide predictive value for investment decisions.

  • We develop a decision support system that enables day-to-day usage of these data for investors.

  • The system supports customized metrics and decision rules based and automatic portfolio management.

  • The portfolios of two test scenarios strongly outperform the market benchmark and comparable public funds, even after risk assessment..

Abstract

Previous research has shown that user-generated stock votes from online communities can be valuable for investment decisions. However, to support investors on a day-to-day basis, there is a need for an efficient support system to facilitate the use of the data and to transform crowd votes into actionable investment opportunities. We propose a decision support system (DSS) design that enables investors to include the crowd's recommendations in their investment decisions and use it to manage a portfolio. A prototype with two test scenarios shows the potential of the system as the portfolios recommended by the system clearly outperform the market benchmark and comparable public funds in the observation period in terms of absolute returns and with respect to the Reward-to-Variability-Ratio.

Section snippets

Introduction and motivation

The rise of user-generated content on the Internet enabled a wider public to participate in online content creation and publication without the need for deep technical expertise. The technical possibility to centrally aggregate the local contributions of a large crowd enables the creation of artifacts which are of equal or superior quality than those made by experts in the domain. Wikipedia, as an example, reaches a comparable quality to the renowned Britannica [19], solely depending on the

Methodology

In this paper we follow the methodology of Design Science Research as suggested by Hevner et al. (2004) which confounds an approach rooted in the engineering sciences. They provide several guidelines which we follow in the development of the proposed architecture [21]:

  • Design as an artifact: Design-science research must produce a viable artifact in the form of a construct, a model, a method, or an instantiation.

  • Problem relevance: The objective of design-science research is to develop

Prototype

In this section, we show implementation details of our prototype system to provide an application example of the proposed design. We show how to setup a simple investment support task, defining the metric and the investment strategy. Subsequently, we run a simple and a more complex test case on a test period of two years and present the results.

Test scenarios

We created two test cases which differ in the way the portfolio is created based on the list of ranked ISINs. Up until the creation of the ranked ISIN list, both test runs are equal.

The first strategy (Test 1) is rather simple, but rational: we invest all cash into the best-ranked ISIN. This is a risky, but also a promising approach. If the forecast is accurate, we maximize our profits by investing in the share with the highest return. We conduct a daily rebalancing of the portfolio: the system

Conclusion

Research shows many approaches for investment decision support systems but none of them enables users to take advantage of the wisdom of crowd even though it has been shown to be advantageous to investment decisions [4], [22] for the price of complex analytical effort. We address this gap and propose the presented system design; it provides flexible means to transform raw crowd vote data into actionable investment decisions, up to automatic portfolio maintenance. By building a prototype and

Acknowledgments

The authors thank Theresa Krimm, Michael Nofer and Renata De Sousa for their comments and support on improving the paper.

Jörg Gottschlich studied business administration and computer science at the University of Bamberg, focusing on information systems, database applications and management and received his diploma degree (equiv. master degree) in 2008. He wrote his diploma thesis about “Continuous improvement of cross-selling offers by automatic analysis of customer reactions” and received the “Excellent Research” audience award at SAS Forum conference in 2008. Since then, Jörg worked for the international

References (43)

  • M. Andrecut

    Portfolio Optimization in R, arXiv

    (2013)
  • W. Antweiler et al.

    Is all that talk just noise? The information content of internet stock message boards

    The Journal of Finance

    (2004)
  • K.J. Arrow et al.

    The promise of prediction markets

    Science (New York, N.Y.)

    (2008)
  • C. Avery et al.

    The “CAPS” Prediction System and Stock Market Returns

    (2011)
  • N. Baba et al.

    Utilization of soft computing techniques for constructing reliable decision support systems for dealing stocks

  • J.E. Berg et al.

    Prediction markets as decision support systems

    Information Systems Frontiers

    (2003)
  • M.M. Carhart

    On persistence in mutual fund performance

    The Journal of Finance

    (1997)
  • L. Chen et al.

    Design and use of preference markets for evaluation of early stage technologies

    Journal of Management Information Systems

    (2009)
  • S. Chou et al.

    A rule-based neural stock trading decision support system

  • B. Claus

    Morningstar Research Report Allianz Thesaurus AT EUR

    (2013)
  • comdirect Bank AG

    Preis- und Leistungsverzeichnis

    (2013)
  • Cited by (0)

    Jörg Gottschlich studied business administration and computer science at the University of Bamberg, focusing on information systems, database applications and management and received his diploma degree (equiv. master degree) in 2008. He wrote his diploma thesis about “Continuous improvement of cross-selling offers by automatic analysis of customer reactions” and received the “Excellent Research” audience award at SAS Forum conference in 2008. Since then, Jörg worked for the international strategy consultancy A.T. Kearney where he provided fact-based decision support for international strategic projects in different industries (e.g. Finance, Retail, and Manufacturing). Jörg joined the Chair of Information Systems at the TU Darmstadt in 2011 as Research Assistant and Ph.D. student. His work includes research on decision support and recommender systems and social networks.

    Oliver Hinz studied at the TU Darmstadt Business Administration and Information Systems with main focus on Marketing, Software Engineering and Computer Graphics. After receiving his diploma (equiv. master degree) he worked several years for the Dresdner Bank as a consultant for business logic. Oliver started working as a Research Assistant in March 2004 at the Chair of Electronic Commerce and received his Ph.D. in October 2007. Oliver Hinz joined the Marshall School of Business (University of Southern California) as visiting scholar for 4 months and received the SinnerSchrader stipend for young researchers for 2007. He has also been awarded with the dissertation prize of the Alcatel-Lucent-Stiftung 2008, the Erich-Gutenberg-Prize 2008 and the science prize “Retailing 2009” of the EHI Retail Institute. He is also the winner of the honorable Schmalenbach prize for young researchers in 2008. He supported the E-Finance Lab as Assistant Professor for E-Finance & Electronic Markets, joined the TU Darmstadt in April 2011 and heads the Chair of Information Systems | Electronic Markets.

    Oliver's research has been published or is forthcoming in journals like Information System Research (ISR), Management Information Systems Quarterly (MISQ), Journal of Marketing (JM), Journal of Management Information Systems (JMIS), International Journal of Electronic Commerce (IJEC), European Journal of Operational Research (EJOR), Decision Support Systems (DSS), Electronic Markets (EM), Business & Information Systems Engineering (BISE) and in a number of proceedings (e.g. ICIS, ECIS, PACIS).

    1

    Tel.: + 49 6151 1675220; fax: + 49 6151 1672220.

    View full text