Papers
Topics
Authors
Recent
Search
2000 character limit reached

Crowds on Wall Street: Extracting Value from Social Investing Platforms

Published 4 Jun 2014 in cs.SI and physics.soc-ph | (1406.1137v1)

Abstract: For decades, the world of financial advisors has been dominated by large investment banks such as Goldman Sachs. In recent years, user-contributed investment services such as SeekingAlpha and StockTwits have grown to millions of users. In this paper, we seek to understand the quality and impact of content on social investment platforms, by empirically analyzing complete datasets of SeekingAlpha articles (9 years) and StockTwits messages (4 years). We develop sentiment analysis tools and correlate contributed content to the historical performance of relevant stocks. While SeekingAlpha articles and StockTwits messages provide minimal correlation to stock performance in aggregate, a subset of authors contribute more valuable (predictive) content. We show that these authors can be identified via both empirical methods or by user interactions, and investments using their analysis significantly outperform broader markets. Finally, we conduct a user survey that sheds light on users views of SeekingAlpha content and stock manipulation.

Citations (16)

Summary

  • The paper demonstrates that targeted sentiment analysis can identify top SeekingAlpha contributors whose insights yield trading returns outperforming the S&P 500.
  • It employs a dictionary-based approach for SeekingAlpha and an SVM model for StockTwits to quantify investment sentiment effectively.
  • By leveraging user engagement heuristics, the study proposes simple long and long/short strategies that highlight practical benefits for investors.

"Crowds on Wall Street: Extracting Value from Social Investing Platforms" Analysis

This paper provides an extensive examination of user-contributed content on social investing platforms, specifically SeekingAlpha and StockTwits. It seeks to determine the quality and predictive power of such content with respect to stock performance and devises methods to leverage valuable insights for trading strategies.

Sentiment Analysis Methodology

The paper employs sentiment analysis as a tool for extracting potentially valuable insights from user-contributed content on these platforms. For SeekingAlpha, a dictionary-based approach was utilized, leveraging a financial sentiment dictionary to calculate sentiment scores based on keyword occurrences. In contrast, due to the short nature of StockTwits messages, a machine learning approach using a support vector machine (SVM) was applied, trained on messages tagged with explicit sentiment labels ("bullish" or "bearish").

Correlation Between Sentiment and Stock Performance

The investigation reveals a general weak correlation between content sentiment and stock price movements. An overall analysis indicates a poor predictive signal from the aggregate content of both platforms. However, the study identifies a subset of contributors on SeekingAlpha whose content shows significantly higher predictive accuracy, thereby emphasizing the existence of valuable analytical content buried within the noise.

Investment Strategy Development

Building on sentiment analysis, the study proposes investment strategies that capitalize on the insights of top-performing SeekingAlpha contributors. Two trading strategies are discussed: a simple "long" strategy and a more aggressive "long/short" strategy. Remarkably, the data demonstrates that by focusing on high-performing contributors, investment strategies based on SeekingAlpha can substantially outperform the S&P 500 index, yielding greater returns while employing straightforward trading methods. This advantage, however, does not translate to StockTwits, where suggested strategies generally underperformed the market baseline.

Heuristics for Identifying Top Contributors

A crucial contribution of the paper is the development of heuristics for identifying top contributors on these platforms. It was found that user engagement metrics, such as comment activity, serve as effective indicators for selecting valuable contributors. Insights gleaned from user interactions proved particularly effective in lieu of manual or empirical evaluation methodologies.

Survey and User Insights

A supplemental survey of SeekingAlpha users and contributors was conducted to assess the perceived impact and trustworthiness of the platform. Findings show high levels of user engagement and reliance on content for making investment decisions, despite prevalent concerns over potential article bias and manipulation intentions. This underscores the need for users to critically evaluate content and signals the importance of community moderation in distinguishing valuable insights from biased contributions.

Conclusion

The study concludes that while social investing platforms like SeekingAlpha and StockTwits host a significant volume of noise, sophisticated sentiment analysis and strategic contributor identification can extract substantive value for investment decisions. This effectively highlights the potential role of crowd wisdom in a domain traditionally dominated by professional financial analysts, suggesting broader applications in leveraging crowdsourcing for expert tasks and decision-making processes. Future work might explore refining sentiment analysis techniques and exploring more complex trading strategies that incorporate diverse data sources and analytical approaches.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Collections

Sign up for free to add this paper to one or more collections.