Are the more popular stocks also the more risky ones?: Google and Wikipedia searches in portfolio optimization
| Název práce v češtině: | Jsou populárnější akcie také ty rizikovější?: Vyhledávání na Google a Wikipedia v optimalizaci portfolia |
|---|---|
| Název v anglickém jazyce: | Are the more popular stocks also the more risky ones?: Google and Wikipedia searches in portfolio optimization |
| Akademický rok vypsání: | 2013/2014 |
| Typ práce: | bakalářská práce |
| Jazyk práce: | angličtina |
| Ústav: | Institut ekonomických studií (23-IES) |
| Vedoucí / školitel: | prof. PhDr. Ladislav Krištoufek, Ph.D. |
| Řešitel: | skrytý - zadáno vedoucím/školitelem |
| Datum přihlášení: | 05.06.2014 |
| Datum zadání: | 06.06.2014 |
| Datum a čas obhajoby: | 16.06.2015 08:00 |
| Místo konání obhajoby: | IES |
| Datum odevzdání elektronické podoby: | 15.05.2015 |
| Datum proběhlé obhajoby: | 16.06.2015 |
| Oponenti: | Mgr. Daniel Benčík |
| Kontrola URKUND: | |
| Předběžná náplň práce |
| In the most recent years, social sciences have obtained access to huge data sets based on internet activity of millions of users all over the world. These massive new data sources can offer a new perspective on the behavior of people. In this thesis we will use the information from Google Trends, a service that shows the popularity of search terms, and Wikipedia in portfolio diversification (the utility of information provided by Google Trends and Wikipedia search queries to the portfolio diversification).
We will propose a portfolio diversification strategy based on the search volume of stock-related terms. The diversification strategy stems in an idea that the more frequently the stock-related term is searched for the higher the risk (in the financial perspective) of a specific stock. According to this assumption, that the popular stocks should be discriminated in the final portfolio, we will assign them lower weights to decrease the total risk of the portfolio. Finally the comparison between search-based portfolios and the constant buy-and-hold strategy will be discussed. The aim of this thesis is to answer the question how can we use data from Google Trends and Wikipedia in portfolio optimization? Can search queries be utilized in portfolio selection and risk diversification? Moreover: How are these queries connected to changes on the stock market? Can movements in trading volume be anticipated by volumes of queries? The correlation between query volumes and trading volumes will be discussed. Hypotheses 1. How does the popularity of the stock contribute to its riskiness? 2. Can trends in financial markets be anticipated from these search queries? 3. Is the search-based strategy of portfolio diversification more valuable than a standard buy-and-hold strategy? 4. Can data from Google Trends predict the fluctuations in stock prices? Methodology The core of the thesis will be based on econometric analysis of search queries provided by Google Trends and Wikipedia. We will use time series analysis and focus on two types of queries: the ticker symbol of a stock and the combination of the word ”stock” and the ticker symbol. We will use our results in portfolio optimization. We will try to analyze as large data sets as possible to confirm our hypothesis; the data go back to January 1, 2004. I will try to introduce a new econometric model suitable for analyzing these data. 1 Core bibliography 1. TSAY, Ruey S. Multivariate time series analysis: with R and financial applications. Hoboken: Wiley,c2014, xiii, 492 s. ISBN 9781118617908. 2. ENDERS, Walter. Applied econometric time series. Hoboken: Wiley, c2010, xiv, 517 s. ISBN 9780470505397. 3. CHAN, Ngai Hang. Time series: applications to finance with R and S-Plus [online]. 2nd ed. Hoboken, N.J.: Wiley, 2010, xxiii, 296 p. 4. RACHEV, S, Stoyan V STOYANOV a Frank J FABOZZI. Advanced stochastic models, risk assessment,and portfolio optimization: the ideal risk, uncertainty, and performance measures [online]. Hoboken, N.J.: Wiley, 2008, xviii, 382 p. 5. Alanyali, M.; Moat, T.; H.S. & Preis. (2013). Quantifying the Relationship Between Financial News andthe Stock Market. Sci.Rep., 3578. doi:10.1038/srep03578 6. Bordino, I., Battiston, S., Caldarelli, G., Cristelli, M., Ukkonen, A., & Weber, I. (2012). Web SearchQueries Can Predict Stock Market Volumes. PLOS ONE 7. doi:10.1371/journal.pone.0040014 7. Choi, H., & Varian, H. (June 2012). Predicting the Present with Google Trends. The Economic Record,vol.88. 8. Kriˇstoufek, L. (2013). Can Google Trends search queries contribute to risk diversification ?Sci.Rep. ,2713. doi:10.1038/srep02713 |