 Recovering Stock Analysts’ Loss Functions from Buy/Sell RecommendationsViola Monostoriné Grolmusz ()
No 2023/4, MNB Working Papers from Magyar Nemzeti Bank (Central Bank of Hungary) Abstract: I carry out an empirical analysis to recover stock analysts’ loss functions from observations on forecasts, actual realizations and a proxy for the publicly observed part of the analyst’s information set. The forecasts I use are analyst stock (buy/hold/sell) recommendations for two Blue Chip stocks. I estimate an asymmetry parameter that captures the analyst’s relative cost from overpredicting versus underpredicting the stock performance. I find that the results are sensitive to the categorization of ‘hold’ recommendations. When substituting ‘holds’ with the recommendation from the previous period, in most cases the estimated bounds for the asymmetry parameter suggest that analysts are more likely to issue a ‘false buy’ than a ‘false sell’ recommenda- tion. This is in line with the frequent statement from the analyst recommendations literature, that optimism relative to the consensus is rewarded in analyst recommendations. By shedding light on the direction of bias in individual analysts’ stock recommendations, we can better understand the operation of financial markets and we can build more accurate models by controlling for these biases. Keywords: Loss functions; Binary forecasting; Preference recovery. (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:mnb:wpaper:2023/4 Access Statistics for this paper More papers in MNB Working Papers from Magyar Nemzeti Bank (Central Bank of Hungary) Contact information at EDIRC. |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.