Work detail

Are the more popular stocks also the more risky ones?: Google and Wikipedia searches in portfolio optimization

Author: Bc. Kristýna Brunová
Year: 2015 - summer
Leaders: prof. PhDr. Ladislav Krištoufek Ph.D.
Work type: Bachelors
Language: English
Pages: 45
Awards and prizes:
Abstract: This thesis studies if the web search data provided by Google Trends and
Wikipedia can be utilized for portfolio diversification. We build up on the
empirical results indicating that the surge in online attention paid towards
a specific stock is associated with an increase in the stock price volatility.
Therefore, we employ a diversification strategy that discriminates for the
popularity of a stock by assigning it a lower portfolio weight. Conversely,
the least searched stocks are preferred in the portfolio. To measure the
popularity of a stock, we focus on Google search volume for stock-related
terms as well as on Wikipedia pageviews of the corresponding company’s
page. Our results show that the search-based strategies outperform the
benchmark index and the uniformly distributed portfolio, reaching lower risk
level and higher standardized average returns. Moreover, these strategies are
successful even in the out-of-sample.




Patria Finance