PhD – Econometrics

DiSES PhD in Economics


Coordinator: Prof. Riccardo LUCCHETTI
Home page: UNIVPM
The econometrics course aims at giving the students all the necessary tools for a thorough comprehension of contemporary applied econometrics literature, plus a solid understanding of the main topics in contemporary econometrics. The course is divided into 3 main modules:

  • General Principles of Estimation and Inference (Riccardo Lucchetti)
  • Time series Econometrics (Giulio Palomba, Riccardo Lucchetti)
  • Microeconometrics (Roberto Esposti, Matteo Picchio)
Module (a) will not go into the specifics of applied econometric techniques, but rather aims at making the students familiar with the main statistical and computational techniques used in contemporary applied econometrics. Modules (b) and (c) are more applied in nature than module (a) and will focus on applications as well as the theoretical aspects.
Econometric methods
Language: English/Italian Frequence: November-December Hours: 40 (16 h. for practice session)
Professor: Riccardo Lucchetti eMail: Web: on DiSES
Teaching Assistant: Claudia Pigini eMail:
Objectives of the Course:
Foundations of modern applied econometric techniques: asymptotic theory (introductory), Maximum Likelihood and Generalized Method of Moments, Robust inference.

  • Asymptotic theory.
  • Maximum Likelihood: Estimation.
  • Maximum Likelihood: Testing.
  • Generalized Method of Moments.
  • Linear Model and Ordinary Least Squares.
  • Instrumental Variable estimation.
  • Robust inference.
Reading List:
Bruce E. Hansen “Econometrics” –
Russel Davidson, James G. Mackinnon “Econometric Theory and Methods”, Oxford University Press.
Software: Gretl.
Time series econometrics
Language: English/Italian Frequence: March-May Hours: 30
Professor: Giulio Palomba, Riccardo Lucchetti eMail:, Web (Palomba): on DiSES
Web (Lucchetti): on DiSES
Objectives of the Course:
  • Time series and stochastic processes.
  • ARMA models.
  • Nonstationary processes and unit roots tests.
  • VAR models.
  • Cointegration.
  • Univariate and multivariate GARCH models.
Reading List:
See for textbooks (in italian).
Hamilton, J.D. (1994), Time series analysis, Princeton University Press.
Software: Gretl (freely dowloadable at
Language: English/Italian Frequence: January-February Hours: 18
Professor: Matteo Picchio eMail: Web: on DiSES
Teaching Assistant: Claudia Pigini eMail:
Objectives of the Course:
  • Binary response models.
    • Linear probability models.
    • Probit models.
    • Logit models.
    • Partial effects and average partial effects.
    • Testing hypothesis in index models.
    • Neglected heterogeneity.
  • Multinomial response models.
    • Multinomial logit models.
    • Partial effects in multinomial logit models.
    • Estimation of multinomial logit models.
    • Independence of irrelevant alternatives.
  • Ordered response models.
    • Ordered logit models.
    • Ordered probit models.
    • Response probabilities in ordered logit and ordered probit models.
    • Log-likelihood function of ordered logit and probit models.
    • Interpretation of estimated parameters.
    • Specification issues.
  • Count response models.
    • Poisson regression.
    • Log-likelihood function in Poisson regression.
    • Efficiency of Poisson quasi-maximum likelihood estimator.
    • Interpretation of estimated parameters.
  • Heteroskedasticity and endogenous regressors in nonlinear models.
    • The heteroskedasticity probit model.
    • Partial effects in heteroskedasticity probit models.
    • Test again heteroskedasticity in probit models.
    • Heteroskedasticity in ordered probit models.
    • Endogeneity in non-linear models.
    • Control function approach in probit models.
    • Testing for endogeneity in probit models.
    • Instrumental variables probit models.
  • The Tobit model.
    • Censored maximum likelihood estimation.
    • Type I Tobit model.
    • Partial effects in Type I Tobit models.
    • Estimation of Type I Tobit models.
    • Specification issues in Type I Tobit models.
  • Sample selection issues.
    • Overview of sample selection.
    • Bivariate sample selection model (Type II Tobit model).
    • Heckman 2-step estimator.
    • Endogenous explanatory variables in Heckit models.
  • Treatment evaluation.
    • Average treatment effect and average treatment effect on the treated.
    • Selection on observables: control function estimator and propensity score matching.
    • Selection on unobservables: difference-in-differences, regression discontinuity design, instrumental variables, timing-of events.
Reading List:
Wooldridge J.M. (2010), Econometric Analysis of Cross Sections and Panel Data, MIT Press.
Cameron A.C. and Trivedi P.K. (2005), Microeconometrics: Methods and Applications, Cambridge.