### Organization

This is the course website for ECON 839, Fall 2016. It will be updated for the next iteration of this course.

• Instructor: Chris Muris
• Office: WMC 3639

• (Class presentations of) your solutions to occasional problem sets (40%)
• Midterm (30%)
• Final (30%)
Dates for ECON 839, Fall 2016
Date Event
Oct 6 No lecture
Oct 18 Midterm
Nov 29 No lecture
Dec 6 Final

### Problem sets

• Problem sets are assigned irregularly
• After you hand in your written solutions, I randomly select one question to be graded.
• At every deadline, I randomly select two students to present their solutions in class.
• If you collaborate or use other sources for your answers, cite those sources and students. If not, I consider it as plagiarism.

For issues such as handing things in late, and on how to deal with illness on midterm/exam day, please see my BUEC 333 syllabus

Readings and slides will be updated as we go.

Readings are from “Econometric Analysis of Cross Section and Panel Data” by Jeffrey W. Wooldridge (second edition). I will abbreviate it by “W”, so that “W2” stands for Chapter 2 in Wooldridge’s book. “W3.1-3.5.1” means “Sections 3.1 up to and including 3.5.1 from Wooldridge”. Note: you should have access to Wooldridge’s book through the SFU library.

The course consists of several blocks, described below. For each block, I estimate the period it will take us to cover it, and indicate required and suggested readings. This information will be updated as we go along, and it is your responsibility to check back frequently.

#### 0. Prereqs

The material that I assume that you are familiar with before coming into this course:

• W1 (Intro)
• W2 (Conditional expectations)
• W3 (Basic asymptotic theory)
• W4 (OLS)
• W5 (IV)

#### 1. Introduction

Introduction to the course. slides and pdf

#### 2. Binary choice + M-estimation

We study several binary choice models, and review and update our understanding of asymptotic theory along the way. slides, pdf, code for maximum score simulations.

Topics:

• logit and probit models
• semiparametric binary choice model
• consistency of M-estimators

• Check that you know the content of W1-W5.
• Lecture notes
• W15.1-15.4, 15.6
• Sections 1-2.6

• for Wooldridge’s notation and CMLE:
• W12.1-12.3
• W13.1-13.5
• W14.1-14.4

#### 3. Latent variable models

We study classic latent variables models of ordered and unordered choice. Then, we look at several structural models. slides, pdf.

Topics

• Ordered choice models
• Unordered choice models
• structural models
• the Roy model
• marriage market model (if there is interest, using Choo and Siow / Galichon and Salanie)
• network models for binary outcomes (if there is interest, using Manski / Brock and Durlauf)

• Latent variables, Wooldridge
• W16.1-16.2.2
• W16.3-16.3.2
• W19.1, 19.3, 19.4, 19.8
• Section 1 of James Heckman’s notes. Skip 1.3.
• Roy model: French and Taber (2011), through page 16, focus on Section 3 (working paper version)

• Roy model:
• Marriage market model
• Peer effects with binary choice

#### 4. Nonparametrics

slides, pdf. A script for NW, LL, and CV

Topics:

• kernel estimation
• local polynomial estimation
• sieve estimation

• Li and Racine (2007), “Nonparametric Econometrics: Theory and Practice”
• Sections 1.1-1.6, 1.9-1.10
• Sections 2.1-2.5

Suggested material:

• Bruce Hansen’s lecture notes
• Chapter 10: up to and including 10.4
• Chapter 11: all, except 11.8 and 11.9
• Chapter 12: 12.1 only
• Hansen (2014), “A Unified Asymptotic Distribution Theory for Parametric and Non-Parametric Least Squares”
• Chen (2007), “Large Sample Sieve Estimation of Semi-Nonparametric Models”, Chapter 76 of the Handbook of Econometrics

#### 5. Semiparametric + inference

Topics:

• two-step M-estimators
• consistency
• asymptotic normality
• nonparametric nuisance parameter
• bootstrap

Required material:

• For consistency and asymptotic normality of two-step M-estimators, study Newey and McFadden
• 3.1-3.3
• 6
• Wooldridge, relevant parts of Chapter 12 (in particular 12.4)
• Bootstrap: lecture notes

Suggested material:

• Two-step:
• Newey (1994), Econometrica, “The asymptotic variance of semiparametric M-estimators”
• Li and Racine, Chapters 7 and 8
• Bootstrap: Horowitz, 2001, Handbook of Econometrics, Chapter 52, “The Bootstrap”

#### 6. Program evaluation

We talk about experiments, methods that work under unconfoundedness (inverse propensity score methods, regression adjustment methods), IV methods, and regression discontinuity design.

• Lecture notes
• Imbens and Wooldridge (2009), “Recent Developments in the Econometrics of program Evaluation.” Journal of Economic Literature 47.1 (2009): 5-86. I am referring to the working paper version below.
• Skip:
• 3.3 and 3.4
• 4.2
• 5.2, 5.6, 5.7, 5.8, 5.12
• read - don’t study - 5.11
• 6.1, 6.2, 6.3.2, 6.5.2, 6.5.3, 6.5.5
• 7
• Wooldridge, Ch. 21
• Blundell and Costa-Diaz (2009), “Alternative Approaches to Evaluation in Empirical Microeconomics” here
• Notes, RDD: NBER lectures by Imbens/Wooldridge
• Notes, unconfoundedness: same lectures
• Arellano’s slides

#### 7. Panel data

We talk about the importance of unobserved heterogeneity and how it can be accommodated in econometric models for repeated measurements. We will start from the fixed effects linear model, and also consider the dynamic model, nonlinear static models with fixed-$$T$$, and recent advances.