Estimating the beliefs of economic agents

Decision making under uncertainty involves a personal belief about the nature of the outcomes. Most economic agents, however, have difficulty gathering information about the future. Despite this difficulty, economists tend to base their insights on models that assume completely informed agents.;This dissertation contributes to the existing literature by studying the expectations of students and their families about the value of education. To achieve this I use both an innovative methodology and new data. Chapters 1 and 2 use a panel survey, the Beginning School Study (1982-2002), with extensive information on the beliefs of a sample of families from the city of Baltimore. I then use this data to show how parents and students update their beliefs on their academic ability and the value of education, from first grade until the end of high school.;Chapter 1 studies how parents and students form their beliefs about the academic grades of their children. I find that families are biased towards expecting higher grades, especially in the case of black parents and parents with boys. Families' ability to predict future scores improved substantially during middle school and high school, with agents' prediction errors having both a lower bias and a lower variance.;In Chapter 2 I show that students in elementary school and their parents believe the wage returns to education are much smaller than the ones estimated by labor economists. Families of lower parental education have particularly low expectations of the wage benefits of education. However, as students age, families update their beliefs about the wage returns to education and their predictions become more accurate.;Graduate studies have become increasingly more important in the American labor market. In Chapter 3 I analyze the expectations of Kellogg MBA students, regarding their future lifetime incomes with and without an MBA degree. I find that MBA students perceive significant income gains and higher returns to work experience from getting an MBA degree.;Overall, my dissertation concludes that students tend to be overoptimistic about their outcomes. I also find that students are very uncertain about outcomes happening several years in the future.