Empirical Finance: Identification Strategies in Corporate Finance
CBS PhD School
Prerequisites
The course is designed as a first-year Ph.D. course. The prerequisites are knowledge of corporate finance theory and econometrics at a M.Sc. level and an ability to work independently with data using a statistical program such as Stata.
Students must participate in the whole course and do all problem sets.
Aim and Objective
The aim of the class is to introduce PhD students in finance and related fields to identification strategies in corporate finance.
The course is designed to provide an introduction to some of the empirical
methods used to identify causal effects in corporate finance.
We will examine how to estimate causal effects in the presence of
potentially unobserved confounding factors and how to make proper
statistical inference about empirical estimates.
The goal of the course is to provide PhD students with a methodological
framework that will enhance their ability to design sound identification
strategies in the area of corporate finance.
Course content
The course is designed to provide an introduction to some of the empirical methods used to identify causal effects in corporate finance. We will examine how to estimate causal effects in the presence of potentially unobserved confounding factors and how to make proper statistical inference about empirical estimates.
The goal of the course is to provide PhD students with a methodological framework that will enhance their ability to design sound identification strategies in the area of corporate finance.
The course content has three main elements:
1. The students will be introduced to the main empirical methods used to identify causal effects in corporate finance. The lectures covers the main econometric challenges as well as guidance on how to estimate causal effects.
2. The course combines lectures on microeconometrics with lectures on seminal papers that apply the empirical methods to research questions in the area of corporate finance.
3. The course has a two problem sets that students must complete.
Teaching style
Lectures
Lecture Plan
Lecture 1: Correlation is not causality: Linear regression models, basic assumptions and causal inference (3 hours)
Lecture 2: Instrumental variables and natural experiments (3 hours)
Lecture 3: Difference-in-differences and panel data (3 hours)
Lecture 4: Regression discontinuity design (2 hours)
Lecture 5: Research design and identification strategies (1 hour)
| Date and time | Topic | Readings: |
|
26/6 9-12 am
Augustinusfonden |
Correlation is not causality: Linear regression models, basic assumptions and causal inference |
Angrist and Pischke, Ch. 1-3. Fazzari, Hubbard, and Petersen (1988) Rajan and Zingales (1998) |
|
26/6 1-4 pm
Augustinusfonden |
Instrumental variables and natural experiments | Angrist and Pischke, Ch. 4. Angrist and Kruger (2001) Bennedsen, Nielsen, Perez-Gonzalez, and Wolfenzon (2007) Nguyen and Nielsen (2010) |
|
27/6 9-12 am
D4.20 |
Difference-in-differences and panel data | Angrist and Pischke, Ch. 5. Yagan (2015) Bertrand and Mullainathan (2003) Bertrand, Duflo, and Mullainathan (2004) |
|
27/6 1-4 pm
D4.20 |
Regression discontinuity design | Angrist and Pischke, Ch. 6. Iliev (2010) Cunat, Gine, and Guadalupe (2012) |
Learning objectives
The course objectives are to:
• Apply microeconometric methods in the area of corporate finance
• Identify econometric challenges in research designs in corporate finance
• Evaluate identification strategies and empirical methods used in corporate finance
• Formulate identification strategies to address econometric challenges in corporate finance
Exam
Please see 'Course content - element 3'
Other
The course is offered through The Nordic Finance Network, and the Department of Finance at CBS will cover the course fee for PhD students from other NFN associated universities.
For further information and registration please follow the link to CBS' course website.