Learning From Data in Economics
Paper Session
Sunday, Jan. 8, 2017 1:00 PM – 3:00 PM
Hyatt Regency Chicago, Grand Ballroom AB
- Chair: James J. Heckman, University of Chicago
What We Have Learned From Structural Models?
Abstract
A structural economic model is one where the structure of decision making is (fully) incorporated in the specification of the model. By identifying the `deep' parameters that describe preferences(technologies) and constraints of the decision-making process, structural models are able to provide counterfactual predictions. In turn they uncover the mechanisms that underpin observed behaviour. Their ability to provide counterfactual policy predictions sets structural models apart from reduced-form models. But structural models require the detailed specification of the decision-making problem - the constraints and the preferences. This paper uses the empirical microeconomic analysis of labor supply and consumer behavior, with some discussion of human capital models as running examples. These are areas where there is a wide spectrum of well-formulated questions from the ex-post impact of past changes in prices, wages and taxes on behaviour, through to the analysis of counterfactual policy changes and optimal design. The focus throughout is on studies that allow a better understanding of the mechanisms underlying observed behaviour and provide reliable insights about policy counterfactuals. Emphasis is given to models that minimize assumptions on the structural function and on unobserved heterogeneity, and approaches that align structural and `reduced form' moments/treatment effects.Updated paper and slides are on my webpage: http://www.ucl.ac.uk/~uctp39a/
Inside Job or Deep Impact? Assessing the Influence of Economics Scholarship
Abstract
TBDAbducting Economics: The Sherlock Holmes Approach
Abstract
TBDDiscussant(s)
Angus Deaton
, Princeton University
Lars Hansen
, University of Chicago
John Rust
, Georgetown University
Burton Singer
, Princeton University
JEL Classifications
- C0 - General