This proposal is a combination of economics and the psychology of decision making. The basic decision making problem studied is repeated decisions made under conditions of risk. How do people evaluate a risky decision or investment decision which is one of a sequence of such decisions? The answer turns on three issues: How they evaluate single trials (or decisions), how they aggregate the trials over time, and how they estimate the cumulative distribution of outcomes. The combination of these three processes is shown to produce myopic loss aversion -- the tendency to overweight the chance of losing on a single trial. This concept of myopic loss aversion is used to try to explain the so-called equity premium puzzle. The equity premium puzzle refers to the fact that stocks have earned much higher returns than bonds in the last century, and the difference is difficult to explain within the standard economic framework. The solution investigated in this project implies that investors evaluate their portfolios annually, even though they usually have a much longer investment horizon. If evaluation horizons were longer, then the equity premium would shrink.
