Micro-econometrics, Causal Inference and Time Series Econometrics

A.Y. 2019/2020
Overall hours
Learning objectives
The aim of this course is twofold. First, to learn how to analyse time-series (typically macro-) data. In particular, how to identify the effect of past shocks on the current state of the world, how to forecast future values and how to model the dynamic interaction between different series.
Second, to analyse the main challenges faced by economists and social scientists in answering empirical questions using micro‐data. The main emphasis will be on learning how to establish causal relationships between different variables and how to use this evidence to inform policy makers' decisions.
Expected learning outcomes
By the end of the course students will be able to:
Understand the difference between a time series and an independent random sample.
Apply non-parametric and parametric techniques to model time series.
Choose and estimate parametric models for time series.
Compute the impulse response function.
Forecast future values.
Handle real‐world data.
Identify causal effects using micro-data
Link econometric theory with data work and produce an insightful and coherent empirical analysis.
Course syllabus and organization

Single session

Lesson period
Second trimester
Prerequisites for admission
There is no formal prerequisite. However, some previous knwoledge of econonometrics is recommended for the micro-econometrics and causal inference module
Assessment methods and Criteria
One written exam for the time series module and one written eam for the micro-econometrics and causal inference module.
Module Micro-econometrics and Causal Inference
Course syllabus
· Introduction: the "credibility revolution"' in empirical economics
· Example: Evaluating Education Policies
· Estimating Causal Policy Effects
· Randomized Experiments
· Regression Discontinuity Design
· Difference in Differences
· Panel Data
· Instrumental Variables
Teaching methods
Theory classes (40 hours)
Teaching Resources
· Lecture slides
· Jousha D. Angrist and Jorn-Steffen Pischke (2015) Mastering Metrics, Princeton University Press
· Specific papers indicated in each chapter
Module Time Series Econometrics
Course syllabus
Topics will include.
Definition of univariate time series.
Non-parametrics characterisation.
Ergodicity and stationarity as generalisation of the iid framework.
Parametric modelling of weakly dependent univariate time series (ARMA modelling). Impulse response functions for ARMA.
Inference in ARMA modelling.
Forecasting with ARMA models.
Model selection: parsimonious modelling.
Parametric modelling of strongly dependent univariate time series: unit root modelling. Forecasting with unit roots.
Unit root testing.
Multivariate modelling: VARMA and VAR modelling for weakly autocorrelated time series. Impulse response function for VARs.
Identification and estimation in VAR models.
Regression models for weakly dependent and unit root multivariate time series (cointegration). Forecasting for multivariate time series.
Large dimensional weakly dependent time series: estimation and forecasting of factor models.
Teaching methods
Theory classes (40 hours)
Teaching Resources
The main reference textbook for the course is:
Time Series Analysis, by J. D. Hamilton, 1994, Princeton University Press .
Other references may be mentioned as the course progresses.
Module Micro-econometrics and Causal Inference
SECS-S/01 - STATISTICS - University credits: 6
Lessons: 40 hours
Professor: Frattini Tommaso
Module Time Series Econometrics
SECS-P/05 - ECONOMETRICS - University credits: 6
Lessons: 40 hours
Professor: Iacone Fabrizio
Please send email to arrange an appointment
Monday 12.30 to 14.30 during term time (Winter term).
Stanza 4 (Second floor)