Use this URL to cite or link to this record in EThOS:
Title: Implications of bias and sentiment in the financial market
Author: Wu, Shan
ISNI:       0000 0004 5191 7353
Awarding Body: University of Exeter
Current Institution: University of Exeter
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
I investigate how career concerns influence banking analysts’ forecasts and find that banking analysts issue relatively more optimistic forecasts early in the year and more pessimistic forecasts later in the year for banks who could be their future employers. This pattern is not observed when the same analysts forecast earnings for banks with no equity research departments. Using the Global Settlement as an exogenous shock on career concerns, I show that this forecast pattern is pronounced after the Settlement. Moreover, I find that analysts benefit from this behaviour as analysts that are more biased in their forecasts towards potential future employers are more likely to move to a higher reputation bank. Textual analysis of analyst reports is also valuable due to the private information and analysis conveyed in the text. Second paper therefore examines analyst reports with consistent and conflicting signals in terms of qualitative and quantitative outputs. I find that investors react more strongly when the sentiment and earnings forecast bias are consistent. Interestingly, when the tone of report text does not coincide with the earnings forecast, investors place greater weight on the text rather than the EPS forecasts. I also find that consistent reports with both optimistic sentiment and forecast bias have a strong positive market reaction but they are low in forecast accuracy. Markedly, forecasts with pessimistic sentiment have higher accuracy than those of optimistic sentiment. Hence, pessimistic sentiment is a good indicator of the quality of forecast reports. Finally, in my last paper, I explore whether there is any association between firm-specific investor sentiment and the subsequent tone of firms' quarterly reports. Firm-specific investor sentiment is measured using the methodology from Aboody et al. (2016), which proxies for market confidence relating to a specific firm. Given the potential cost-benefit trade-off in the reporting strategy, I argue and find different responses from managers in their 10-Qs in terms of their investor sentiment. I focus on the tone of optimism, readability and the proportion of uncertain words in the 10-Q filings. For firms with extremely high levels of investor sentiment, managers tend to be more conservative by using less optimistic words to avoid future disappointment. In comparison, in firms with extremely pessimistic investor sentiment, managers tend to use more optimistic and easy to understand language, and minimize their proportion of uncertainty in their 10-Q filings. By doing so, perhaps they are trying to alter their investor sentiment.
Supervisor: Gyoshev, Stanley Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: analyst forecast ; investor reaction ; conflicting signals ; investor sentiment ; textual analysis