skip to main content
Ec 123
Econometric Analysis of Discrete Choice
9 units (3-0-6)  | second term
Prerequisites: Ec 122.
This course uses advanced econometric tools to analyze why people make the choices that they do in many domains- whether to make investments in peer-to-peer lending, consumer shopping between brands, where to go to college (Caltech or MIT?), choosing between modes of transportation (car, metro, Uber/Lyft, or bicycle), etc. We will focus on applications of discrete choice models (in which the dependent variable to be explained is usually a 0-1, Yes or No choice). The statistical models create estimates of behavioral parameters which describe numerically how much people value different goods or outcomes and how random their responses are. Models studied include logit, nested logit, probit, and mixed logit etc. Simulation techniques that allow estimation of otherwise intractable models will also be discussed. Models of this kind are routinely used in business and government, but are often misused and misinterpreted unless they are deeply understood.
Instructor: Xin