Meetings: 2011 Italian Stata Users Group meeting

TStat, Stata’s distributor in Italy, is pleased to announce that the 8th Italian Stata Users Group meeting will take place in Venice on November 17–18, 2011. The meeting provides Stata users working in different research areas with a unique opportunity to exchange ideas, experiences, and information on user-written routines and applications. Stata users interested in contributing to the meeting are encouraged to submit their proposals to the scientific committee. As in previous years, the emphasis will be on the development of new commands or procedures currently unavailable in Stata.

Registration

Preliminary program and training courses

The deadline for submissions has passed. A preliminary selection will be made, on the basis of submitted abstracts, by the scientific committee by September 7, 2011. The final version of the paper must be submitted to the conference organizer by October 31, 2011.

Thursday, November 17: Presentations

9:00–9:20	Registration of participants / Welcome from the scientific committee
9:30–10:45	Session I — Invited speaker
10:45–11:00	Coffee break
11:00–12:45	Session II — User-written commands and routines I
12:45–2:15	Lunch
2:15–3:30	Session III — Exploiting the potential of Stata
	Multiple imputation using Stata’s mi command Yulia Marchenko StataCorp
3:30–4:45	Session IV — User-written commands and routines II
4:45–5:00	Coffee break
5:00–5:30	Session V — Report to users / Wishes and grumbles
	David Drukker StataCorp
6:00	Guided tour of the island
8:00	Optional dinner at Vecio Fritolin

Friday, November 18: Training courses

Introduzione all’analisi dei dati multivariati con Stata (in italiano)

Obiettivo: Il corso si propone di fornire ai partecipanti un’introduzione ai metodi per l’analisi di dati multivariati attraverso l’impiego di Stata. Grazie all'enorme quantità di dati ormai disponibili in ogni settore industriale e commerciale, le tecniche di analisi statistica multivariata ricoprono oggi più mai un ruolo fondamentale per l’estrazione di utili informazioni dai dati stessi. Durante il corso saranno illustrate le principali metodologie di analisi multivariata (analisi dei cluster, analisi delle componenti principali, analisi fattoriale) attraverso esempi e casi concreti.

Destinatari: Il corso è di interesse per ricercatori/analisti che desiderino condurre ricerche empiriche utilizzando dati multivariati.

Prerequisiti: Conoscenza di base di statistica e del software Stata.

Sessione I (9:00–11:00) — I dati multivariati: Primi indicatori di sintesi e rappresentazioni grafiche

Tipi di variabili e il problema dei dati mancanti
Covarianze, correlazioni e misure di distanza
La distribuzione normale multivariata
Grafici per la visualizzazione di dati multivariati

Sessione II (11:15–1:00) — Analisi delle componenti principali e analisi fattoriale

Analisi delle componenti principali
- Introduzione
- Calcolo delle componenti principali
- Calcolo degli scores delle componenti principali
Analisi dei fattori
- Introduzione
- Stima dei fattori
- Scelta del numero di fattori
- Rotazione dei fattori

Sessione III (2:00–3:45) — Analisi dei cluster

Introduzione agli algoritmi di clustering
Principali metodi agglomerativi di clustering
- Il dendrogramma
- Single linkage
- Complete linkage
- Average linkage
- Metodo di Ward
Principali metodi divisivi di clustering
- K-means
Profilazione dei cluster

Sessione IV (4:00–5:00) — Altri metodi di analisi dei dati multivariati

Estensioni degli approcci presentati
Altre tecniche di analisi multivariata (analisi delle corrispondenze, scaling multidimensionale, etc.)

Numero massimo di iscritti: 20
Termine iscrizione: 5 Novembre 2011

Treatments effects estimation using Stata (in English)

Presenter: David Drukker, StataCorp

Course objectives: To provide an introduction to estimators of a binary treatment effect when individuals are selected into treatment based observable variables and some randomness that is not related to any observable variable used to model the response. After providing an introduction to the potential outcome framework, we will discuss some parametric models for treatment effects estimation to gain some intuition. Then we will progressively drop many of the strong assumptions in the original parametric model and use semiparametric and nonparametric methods to estimate the treatment effect. During the course, we cover matching estimators, inverse propensity-score weighting estimators, and series estimators.

Potential audience: Researchers interested in using Stata for treatment effects estimation

Prerequisites: Familiarity with some Stata estimation commands and the English language

Session I

An introduction to the potential outcome framework
The meaning of “selection on observables”
Other identifying assumptions

Session II

Some simple parametric estimators
The estimation and use of the propensity score
Some series estimators

Session III

The Imbens–Abadie matching estimators

Session IV

Some double-robust estimators

References:

Abadie, A. and G. W. Imbens. 2006.: Large sample properties of matching estimators for average treatment effects. Econometrica: 74: 235–267.

Abadie, A. and G. W. Imbens. 2011.: Bias-corrected matching estimators for average treatment effects. Journal of Business and Economic Statistics: 29, 1–11.

Imbens, G. W. and J. M. Wooldridge. 2009.: Recent Developments in the Econometrics of Program Evaluation. Journal of Economic Literature 47: 5–86.

Wooldridge, J. M. 2010.: Econometric Analysis of Cross Section and Panel Data. 2nd ed. chap. 21. Cambridge, MA: MIT Press.

Maximum number of participants: 20
Registration Deadline: November 5, 2011

Organizer

The meeting is being organized by TStat S.r.l. (www.tstat.it), the distributor of Stata in Italy.

For more information or assistance with accommodations, please contact statausers@tstat.it.

2011 Italian Stata Users Group meeting

Announcement and call for presentations

Registration

Preliminary program and training courses

Thursday, November 17: Presentations

Friday, November 18: Training courses

Organizer