This thesis consists of two independent essays which make use of economic experiments in order to understand people’s behaviors in the financial market and their time-related choices.

The first essay uses the WeChat platform, one of the largest social networks, to conduct an online experiment of artificial investment games. We investigate how people’s forecasts about the financial market and investment decisions are shaped by whether they can observe others’ forecasts and whether they engage in public or private investment decisions. We observe a strong positive correlation between forecasts and investments, suggesting an association between the subjects’ reported forecasts and their beliefs. More importantly, we find that with forecast sharing, subjects’ forecasts converge but...[ Read more ]

This thesis consists of two independent essays which make use of economic experiments in order to understand people’s behaviors in the financial market and their time-related choices.

The first essay uses the WeChat platform, one of the largest social networks, to conduct an online experiment of artificial investment games. We investigate how people’s forecasts about the financial market and investment decisions are shaped by whether they can observe others’ forecasts and whether they engage in public or private investment decisions. We observe a strong positive correlation between forecasts and investments, suggesting an association between the subjects’ reported forecasts and their beliefs. More importantly, we find that with forecast sharing, subjects’ forecasts converge but in different directions across groups; consequently, forecast sharing does not lead to better forecasts nor better investment performance. Whether or not subjects engage in public investment decisions does not significantly affect forecasts or investment.

The second essay points out a potential gap between intertemporal choices and time preference: the observed intertemporal decisions could be driven by a biased perception of time, and thus may not completely reveal the actual time preference. To test this conjecture, we explore the relationship between time perception and the observed intertemporal choices by conducting a laboratory experiment, in which cognitive load is used as an instrument to manipulate time perception. We find that the perceived time lengths for subjects with high cognitive load are shorter than those with low load. We also observe a correlation between an individual’s time overestimation and his apparent impatience. Our key result suggests that time perception indeed mediates part of the cognitive load’s effect on intertemporal choices, supporting our conjecture.

Keywords: economic experiment, forecast, investment, time preference, time discounting