- Economics For Data Science M
- Summary
Course Syllabus
Obiettivi formativi
Al termine del corso, gli studenti avranno imparato come:
· Costruire un modello economico per identificare e stimare un impatto causale.
· Confrontare e valutare diverse strategie di econometria applicata per stimare un impatto causale.
· Usare big data e machine learning per la stima di un impatto causale.
·
Capire vantaggi e valore aggiunto dell'utilizzo di big data per ricerca applicata nelle scienze sociali.
Al termine del corso, gli studenti sapranno:
· Utilizzare le tecniche di program evaluation per rispondere a domande di ricerca che pongono problemi di policy-relevance.
· Utilizzare in modo efficacie i big data per rispondere a importante domande di ricerca economica.
Contenuti sintetici
Questo corso introduce il field emergente che nasce dalla fusione di Economia e Data Science per rispondere a domande di policy-relevance. L'obiettivo principale del corso e' discutere le piu' importanti tecniche di machine learning e di econometria applicata per stimare un impatto causale, e i vantaggi dell'utilizzo di big data per rispondere a importanti domande di ricerca in diverse applicazioni.
Discuteremo i seguenti quattro argomenti principali:
1) Validta' del Modello e Inferenza Causale.
2) Machine
Learning e Inferenza Causale.
3) Modelli Strutturali, Esperimenti e Machine
Learning.
4) Applicazioni Empiriche che usano Big Data.
Programma esteso
Topic 1: Model’s Validity and Causal Inference.
- Internal and external validity
- Big data: new frontiers for economic analysis
Topic 2: Machine Learning and Causal Inference.
- Program evaluation and the “missing” counterfactuals
- Causal inference and machine learning
Topic 3: Structural Models, Experiments and Machine Learning
- Developing, estimating and using structural models
- Comparing experiments, structural models and machine learning methods for causal inference
Topic 4: Empirical Applications Using Big Data
- Students’ presentations: present your own work, one of the papers from a list of suggested papers that will be provided, or a paper of your choice that uses machine learning methods, possibly replicating the results of the paper you choose to present.
Prerequisiti
Principi di econometria applicata e metodi quantitativi di statistica applicata.
Metodi didattici
Lezioni e progetti degli studenti (individuali o di gruppo).
Modalità di verifica dell'apprendimento
Presentazioni e progetti di applicazione dei modelli e metodi economici ai dati.
Testi di riferimento
Libri di testo: per questo corso non c'e' un libro di testo di riferimento. Di seguito un elenco di alcuni testi di riferimento per gli argomenti di econometria trattati nel corso. Tutti i libri di
testo di seguito riportati sono disponibili in formato e-book tranne Wooldridge (2020) che
e' disponibile presso la Biblioteca di Ateneo sede Centrale e Sede di
Scienze.
· W. H. Greene. Econometric Analysis, 5th Edition, Prentice Hall International, 2003.
Semplice/meno tecnico:
· J. Wooldridge. Introductory Econometrics: A Modern Approach, 7th Edition, Cengage Learning, 2020. (for IV and 2 stage least squares)
· Stock and Watson, Introduction to Econometrics, 3rd Edition. (Basic statistics and regression analysis; companion website with datasets and files to replicate empirical results: https://wps.pearsoned.com/aw_stock_ie_3/178/45691/11696965.cw/index.html)
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009. (Excellent for concept of causality, experiments, diff-in-diff, and RD)
Articoli e capitoli di libro: la discussione di ciascuno dei quattro argomenti di cui discuteremo nel corso fara' riferimento ai seguenti principali articoli:
Topic
1: Model’s Validity and Causal Inference.
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009, Chapter 1 pages 3-8.
· Athey, Susan, 2017. “Beyond prediction: Using big data for policy problems”, Science, 355, 483-485.
· Einav L., and J. Levin. 2014. “Economics in the Age of Big Data”, Science, Vol 346, Issue 6210: 1243089.
· Kleinberg, John, Jens Ludwig, Sendhil Mullainathan, and Ziad Obermeyer. 2015. “Prediction Policy Problems” American Economic Review, 105(5): 491–95.
· Kleinberg, John, Jens Ludwig, and Sendhil Mullainathan. 2016. “A Guide to Solving Social Problems with Machine Learning”, Harvard Business Review.
· Shmueli, G. 2010. “To Explain or to Predict?”, Statistical Science, Vol. 25, No. 3, 289-310.
· Varian, H. 2014. “Big Data: New Tricks for Econometrics”, Journal of Economic Perspective 28, 3-28
Topic 2: Machine Learning and Causal Inference.
· Athey S. and G. Imbens. 2015. “Machine Leaning Methods in Economics and Econometrics”, American Economic Review: Papers and Proceedings, 105(5): 476-480.
· Varian, H. (2016) “Causal Inference in Economics and Marketing”, PNAS, Vol. 113, No. 27, pages 7310-7315.
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009, Chapter 2 pages 11-24, Chapter 6 pages 251-267.
Topic 3: Structural Models, Experiments and Machine Learning.
· Attanasio O., Costas Meghir and Ana Santiago. 2011. "Education choices in Mexico: using a structural model and a randomised experiment to evaluate Progresa", Review of Economic Studies.
· Todd, Petra, and Kenneth, I. Wolpin. 2006. "Assessing the Impact of a School Subsidy Program in Mexico: Using a Social Experiment to Validate a Dynamic Behavioural Model of Child Schooling and Fertility”, American Economic Review, Vol. 96, Issue 5, pages 1384-1417.
Topic 4: Empirical Applications Using Big Data.
References: a list of papers will be provided.
Periodo di erogazione dell’insegnamento
Secondo semestre.
Lingua di insegnamento
Inglese.
Learning objectives
At the end of the course, you will learn how to:
· Build an economic model to identify and estimate a causal effect.
· Compare and assess alternative applied econometrics strategies to estimate a causal effect.
· Use big data and machine learning for causal inference.
· Understand the advantages and value added of using big data for applied research in social sciences.
At the end of the course, you will be able to:
· Be familiar with the most important approaches to program evaluation to address a variety of policy-relevant research questions.
· Effectively use big data to address important research questions in Economics.
Contents
This course introduces the emerging field
that merges Economics and Data Science to answer policy relevant research
questions. The main goal of the course is to discuss the most important machine
learning and applied econometrics techniques to estimate a causal effect, and
the advantages of using big data to answer relevant research questions in
several applications.
We will discuss four main topics:
1) Model’s Validity and Causal Inference.
2) Machine
Learning and Causal Inference.
3) Structural Models, Experiments and Machine
Learning.
4) Empirical Applications Using Big Data.
Detailed program
Topic 1: Model’s Validity and Causal Inference.
- Internal and external validity
- Big data: new frontiers for economic analysis
Topic 2: Machine Learning and Causal Inference.
- Program evaluation and the “missing” counterfactuals
- Causal inference and machine learning
Topic 3: Structural Models, Experiments and Machine Learning
- Developing, estimating and using structural models
- Comparing experiments, structural models and machine learning methods for causal inference
Topic 4: Empirical Applications Using Big Data
- Students’ presentations: present your own work, one of the papers from a list of suggested papers that will be provided, or a paper of your choice that uses machine learning methods, possibly replicating the results of the paper you choose to present.
Prerequisites
Principles of applied econometrics and statistical quantitative methods for data analysis.
Teaching methods
Lectures and students' projects (individual or group).
Assessment methods
Textbooks and Reading Materials
Textbooks: there is no given recommended textbook for this course. Below a list of some Econometrics textbooks for reference. All listed textbooks are available as e-books with the exception of Wooldridge (2020), which is available at the University Library (both Central Site and Science Site).
Advanced:
· W. H. Greene. Econometric Analysis, 5th Edition, Prentice Hall International, 2003.
Simpler/less math:
· J. Wooldridge. Introductory Econometrics: A Modern Approach, 7th Edition, Cengage Learning, 2020. (for IV and 2 stage least squares)
· Stock and Watson, Introduction to Econometrics, 3rd Edition. (Basic statistics and regression analysis; companion website with datasets and files to replicate empirical results: https://wps.pearsoned.com/aw_stock_ie_3/178/45691/11696965.cw/index.html)
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009. (Excellent for concept of causality, experiments, diff-in-diff, and RD)
Journal articles and book chapters: the discussion of each the four topics that we will discuss will make reference to the following main articles.
Topic
1: Model’s Validity and Causal Inference.
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009, Chapter 1 pages 3-8.
· Athey, Susan, 2017. “Beyond prediction: Using big data for policy problems”, Science, 355, 483-485.
· Einav L., and J. Levin. 2014. “Economics in the Age of Big Data”, Science, Vol 346, Issue 6210: 1243089.
· Kleinberg, John, Jens Ludwig, Sendhil Mullainathan, and Ziad Obermeyer. 2015. “Prediction Policy Problems” American Economic Review, 105(5): 491–95.
· Kleinberg, John, Jens Ludwig, and Sendhil Mullainathan. 2016. “A Guide to Solving Social Problems with Machine Learning”, Harvard Business Review.
· Shmueli, G. 2010. “To Explain or to Predict?”, Statistical Science, Vol. 25, No. 3, 289-310.
· Varian, H. 2014. “Big Data: New Tricks for Econometrics”, Journal of Economic Perspective 28, 3-28
Topic 2: Machine Learning and Causal Inference.
· Athey S. and G. Imbens. 2015. “Machine Leaning Methods in Economics and Econometrics”, American Economic Review: Papers and Proceedings, 105(5): 476-480.
· Varian, H. (2016) “Causal Inference in Economics and Marketing”, PNAS, Vol. 113, No. 27, pages 7310-7315.
· Angrist and Pischke, Mostly Harmless Econometrics, Princeton and Oxford University Press, 2009, Chapter 2 pages 11-24, Chapter 6 pages 251-267.
Topic 3: Structural Models, Experiments and Machine Learning.
· Attanasio O., Costas Meghir and Ana Santiago. 2011. "Education choices in Mexico: using a structural model and a randomised experiment to evaluate Progresa", Review of Economic Studies.
· Todd, Petra, and Kenneth, I. Wolpin. 2006. "Assessing the Impact of a School Subsidy Program in Mexico: Using a Social Experiment to Validate a Dynamic Behavioural Model of Child Schooling and Fertility”, American Economic Review, Vol. 96, Issue 5, pages 1384-1417.
Topic 4: Empirical Applications Using Big Data.
References: a list of papers will be provided.
Semester
Second semester.
Teaching language
English.
Key information
Staff
-
Chiara Binelli