We explore the mental models of the stock market among different groups of economic agents.
Experimental design
In our surveys, we present participants with different hypothetical scenarios describing news about the future earnings stream of a company. We explore whether investors still believe that the old news are relevant for future returns and why they think or do not think so. Moreover, we explore how investors' beliefs differ across the different investor classes.
Design details
Randomization method
Participants will be randomly assigned to different hypothetical news cases. The randomization is computerized in Qualtrics.
Randomization unit
Individual
Clusters
Equal to number of observations.
Observations
General population (US), Dynata, n = 2,400
Financial advisors (US), CloudResearch, n = 200
Financial professionals (US), CloudResearch, n = 200
Academic experts (global), invited via email, n = 150*
General population (Germany), Bundesbank Online Panel, 4,350**
*Sample size is approximate/aspirational because we cannot perfectly predict the response rate.
**Sample size estimated by Bundesbank Online Panel team.
Arms
General population (US): All prediction conditions (6 individual stocks, 4 aggregate cases). 200 respondents per condition, except for the two Nike cases, for which we plan to collect 400 respondents each.
Financial advisor / financial professionals / academic experts: Randomly assigned to the Nike good news or the Nike bad news case with equal chance.
General population (Germany): All six individual stock prediction conditions. Condition selected randomly with equal chance.
Primary outcomes
- investors’ return expectations
- the reasoning behind their expectation
Cited by
No citation data available for this registration yet.