Sharing investment ideas: Role of luck and skill

• Název práce v češtině: Sdílení investičních nápadu: Rola štěstí a dovednosti
• Název v anglickém jazyce: Sharing investment ideas: Role of luck and skill
• Klíčová slova: štěstí vs. dovednosti
• Klíčová slova anglicky: skill vs. luck, multiple hypothesis testing, public information sharing
• Akademický rok vypsání: 2020/2021
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: Institut ekonomických studií (23-IES)
• Vedoucí / školitel: Mgr. Martin Hronec
• Řešitel: skrytý - zadáno vedoucím/školitelem
• Datum přihlášení: 22.07.2021
• Datum zadání: 22.07.2021
• Datum a čas obhajoby: 15.09.2021 09:00
• Místo konání obhajoby: Opletalova - Opletalova 26, O314, Opletalova - místn. č. 314
• Datum odevzdání elektronické podoby: 27.07.2021
• Datum proběhlé obhajoby: 15.09.2021
• Oponenti: doc. Bc. Jiří Novák, M.Sc., Ph.D.

Seznam odborné literatury

Barras, Laurent, O Scaillet, and Russ Wermers. “False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas,” 2010, 53.

Barras, Laurent, O Scaillet, and Russ Wermers. “Reassessing False Discoveries in Mutual Fund Performance: Skill, Luck, or Lack of Power? A Reply,” 2019, 37.

Chen, Yong, Michael Cliff, and Haibei Zhao. “Hedge Funds: The Good, the Bad, and the Lucky.” Journal of Financial and Quantitative Analysis 52, no. 3 (June 2017): 1081–1109.

Crawford, Steven S., Wesley R. Gray, and Andrew E. Kern. “Why Do Fund Managers Identify and Share Profitable Ideas?” Journal of Financial and Quantitative Analysis 52, no. 5 (October 2017): 1903–26.

Fama, Eugene F, and Kenneth R French. “Luck versus Skill in the Cross-Section of Mutual Fund Returns,” 2010, 33.

Ferson, Wayne and Chen, Yong, (2020), How Many Good and Bad Funds are There, Really?, ch. 108, p. 3753-3827 in , HANDBOOK OF FINANCIAL ECONOMETRICS, MATHEMATICS, STATISTICS, AND MACHINE LEARNING, World Scientific Publishing Co. Pte. Ltd..

Harvey, Campbell R., and Yan Liu. “Luck versus Skill in the Cross-Section of Mutual Fund Returns: Reexamining the Evidence.” SSRN Electronic Journal, 2020.

Harvey, Campbell R, and Yan Liu. “Detecting Repeatable Performance.” The Review of Financial Studies 31, no. 7 (July 1, 2018): 2499–2552.

Storey, John D. “A Direct Approach to False Discovery Rates.” Journal of the Royal Statistical Society: Series B (Statistical Methodology) 64, no. 3 (August 2002): 479–98.

Předběžná náplň práce v anglickém jazyce

With the widespread availability of the internet came also the widespread availability of the investment recommendations. It is in the best interest of individual investor to determine which recommendations they should follow, which is however not so easy to do. There are thousands of people-analysts who post their ideas on the internet and just by the pure luck alone, there must be some number of them which exhibit abnormal positive past returns.

To study this problem, we however cannot just observe any group of analysts, as we should keep at least some minimum requirements on the professionalism of the given analyst. This is to assure that they are really not just randomly picking stocks and also that the research has some useful value for the practice. For these reasons we study the data of the closed, but public, analyst group, which has some entry requirements for joining.

The pool of studies on the performance of individual analysts is large. For our setting of public information sharing Xi (2005) show that simply following ideas of the analysts who have above median prediction skills in the past period can generate positive abnormal returns in the future. However, Hall (2010) do not find the link between past performance and the future results. Andreas (2013) try to identify skilled analysts by using their own measure of relative forecast accuracy, which controls for the forecast complexity. Using this measure, they are able to outperform the market returns.
To separate the skillful analysts from the just lucky ones, we need to however introduce multiple hypothesis testing (MHT) as the simple hypothesis testing of each person’s ideas does not take into account that we are testing thousands of them. These studies however do not apply multiple hypothesis testing and are therefore prone to the selection bias under multiple testing.
However, as the literature analyzing performance of the mutual/hedge funds often incorporates MHT, it should be a good idea to look into it, as our area is actually not that far from the fund setting.
One of the important works in this area is Barras et al. (2010), which integrates the False Discovery Ratio (FDR) approach from Storey (2002), into the financial setting. Their results suggest that there is a substantial, but small, group of funds which outperform the market. They also find that the number of those funds is getting rapidly smaller throughout the time.
Another influential paper is Fama & French (2010), who by using bootstrapping find that only a few mutual funds outperform the market, which Harvey et al. (2020) contribute to the too conservative approach, caused by the low power of their tests.
Building upon, Barras et al. (2010), there are various extensions to the original FDR method. Just to mention, Chen et al. (2017) introduce parametrical approach to FDR. This approach also allows for the ranking of performance of funds, which classical FDR fund doesn’t allow as it only acts as a threshold.