- A review of the classical linear regression model
- Key panel data techniques
- An overview of the matching models and regression discontinuity designs
- Computer exercises using the statistical software package “Stata” are an integral part of the course, which ensures that the students get hands-on experience of analyzing real world data
- Know key methods of econometric research, understand the causes and consequences of endogeneity, know the main methods for addressing this problem
- Understand endogeneity as a key issue affecting causal inference; be able to critically examine existing research from this angle
- Be able to apply the methods learnt when conducting own empirical analysis
- Be familiar with and be able to use key capabilities of the statistical package “Stata”, including its programming options (the so-called do-files)
- Understand the limits of interpreting regression results in most settings (the ceteris paribus clause).
- Overview of the classical linear regression modelL1.1. The classical linear regression model. OLS estimation. L1.2. Inference in the CLRM. L1.3. OLS asymptotics. L1.4. Specification and data issues. Reading: Wooldridge (2016), chapters 3-7; Hansen (2017), chapter 4, 7; Lecture notes.
- Introduction to econometric package StataCL1.1. Basic capabilities of Stata. Basic commands. Do and log files. CL1.2. The grammar of Stata. CL1.3. Creating and changing variables in Stata. Reading: Stata manual (2015); Lecture notes.
- Endogeneity. Instrumental variables methodsCL2.1. Key commands of regression analysis. Hypothesis testing and model diagnostics. Reading: Stata manual (2015); Lecture notes.
- Analysis of panel (longitudinal) dataL3.1. Examples of panel data. L3.2. Fixed and random effects models. L3.3. Model diagnostics (the Hausman test, etc.). L3.4. Two-way fixed effects models. L3.5. Endogenous explanatory variables. L3.6. The Hausman-Taylor model. L3.7. Dynamic panel data models. Reading: Wooldridge (2016), chapters 13-14.CL4.1. Fixed- and random-effects models in Stata. CL4.2. Model diagnostic (the Hausman test, etc.). CL4.3. The Hausman-Taylor model. CL4.4. Dynamic panel data models. Reading: Stata manual (2015); Lecture notes.
- Estimation of treatment effects. The difference-in-difference estimatorL4.1. Statistical setup. Selection on observables and selection on unobservables. Characterizing selection bias. L4.2. The difference estimators and the DiD. L4.3. Testing the key assumption of the DiD. Reading: Cerulli (2015), chapter 1, 3.4; Roberts and Whited (2013), chapter 4 (стр. 520-531).
- Propensity score matching and regression discontinuity models5.1. Matching models. Treatment effects and necessary identifying assumptions. Propensity score matching. 5.2. Regression discontinuity (RD) models. Sharp and fuzzy regression discontinuity designs. Identification of treatment effects in the sharp RD. Reading: Cerulli (2015), chapter 2.3 and 4.3; Roberts and Whited (2013), chapters 5 (pp. 531-549) and 6 (pp. 549-557).
- Four home assignments
- Mid-term test
- Final examThe duration of the final exam is two academic hours (80 mins). on-line
- Interim assessment (3 module)0.5 * Final exam + 0.3 * Four home assignments + 0.2 * Mid-term test
- Giovanni Cerulli. (2015). Econometric Evaluation of Socio-Economic Programs. Springer. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsrep&AN=edsrep.b.spr.adstae.978.3.662.46405.2
- Atanasov, V., & Black, B. (2016). Shock-Based Causal Inference in Corporate Finance and Accounting Research. Critical Finance Review, (2), 207. https://doi.org/10.1561/104.00000036
- Bruce E. Hansen. (2013). Econometrics. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.C0DB9E1E
- Roberts, M. R., & Whited, T. M. (2013). Endogeneity in Empirical Corporate Finance1. Handbook of the Economics of Finance, 493. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsrep&AN=edsrep.h.eee.finchp.2.a.493.572