TStat - O F F I C I A L

NOVITA'/CONFERENZE E INCONTRI

5� CONVEGNO ITALIANO DEGLI UTENTI DI STATA - Milano, 20-21 ottobre 2008

TStat S.r.l., distributrice esclusiva del software Stata in Italia, � lieta di annunciare la quinta riunione italiana degli utenti di Stata, che si terr� il 20-21 ottobre 2008 presso il Jolly Hotel Machiavelli, Via Lazzaretto 5, Milano. L’incontro � aperto a tutti gli utenti di Stata. Il modulo di registrazione ed eventuali altre informazioni possono essere ottenuti da Paola di Rienzo, tel. 0864 210101, e-mail: statausers@tstat.it.

PROGRAMMA

Luned�, 20 ottobre 2008

8.45 - 9.15 Registrazione dei partecipanti

9.15 - 10.30 I SESSIONE – INVITED SPEAKER – CHAIR Marcello Pagano

Latent variable and other methods for cohort and other cross-sample/cross measure comparisons, Andrew Pickles (Health Methodology Research Group, University of Manchester, England.)

10.30 - 10.45 Pausa caff�

10.45 - 12.15 II SESSIONE - USER WRITTEN COMMANDS AND ROUTINES I - CHAIR Una-Louise Bell

Confidence Bands for the Survival Function – Enzo Coviello (Unit� di Statistica ed Epidemiologia ASL BA/2, Giovinazzo)
New wine in new bottles: visualizing the progression over time of the epidemics of tobacco smoking and obesity through the use of modified population pyramids – Giovanni Capelli e Bruno Federico (Dipartimento di Scienze Motorie e della Salute, Universit� di Cassino) e Giuseppe Costa (Departmento di Sanit� Pubblica e Microbiologia, Universit� di Torino)
Ordered Probit Models with Anchoring Vignettes - Claudio Rossetti (University of Rome, “Tor Vergata”)

12.15 - 13.15 III SESSIONE - EXPLOITING THE POTENTIAL OF STATA 10 - CHAIR Rino Bellocco

Reproducible Research - Weaving with Stata – Bill Rising (StataCorp)
Esempio utilizzato durante la presentazione (formato ZIP)

13.15 - 14.30 Pranzo

14.30 - 16.30 IV SESSIONE - USER WRITTEN COMMANDS AND ROUTINES II - CHAIR Rino Bellocco

Parametric and semiparametric estimation of ordered response models with sample selection and individual-specific thresholds - Giuseppe de Luca and Valeria Perotti (ISFOL) and Claudio Rossetti (University of Rome, “Tor Vergata")
A Simulation-Based Sensitivity Analysis for Matching Estimators - Tommaso Nannicini (Universidad Carlos III, Madrid)
Estimating and testing multi-way error component models with correlated effects in Stata, Giovanni Bruno - Universit� Bocconi, Milano)

A seasonal unit root test with Stata - Domenico Depalo (Universit� di Roma, “Tor Vergata”)

16.30 - 16.45 Pausa caff�

16.45 - 17.45 V SESSIONE - TEACHING WITH STATA - CHAIR Giovanni Capelli

The use of Stata in Biostat Teaching - Rino Bellocco (Dipartimento di Statistica, Universit� di Milano – Bicocca e Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden)
Discussione – Marcello Pagano (Department of Biostatistics, Harvard School of Public Health) e Maurizio Pisati (Dipartimento di Sociologia e Ricerca Sociale, Facolt� di Sociologia, Universit� di Milano – Bicocca)

17-45 - 18.15 REPORTS TO USERS / WISHES AND GRUMBLES

Alan Riley (StataCorp)

Marted�, 21 ottobre 2008

CORSI DI FORMAZIONE

9.00 - 17.00 - Multilevel Modeling in Stata (in Inglese)

9.00 - 10.50 Sessione I - Fixed and Random Effect Estimates

Non-independent sampling, matching and assumptions about confounding

Within and between estimators

Random intercept models

xtreg, xtmixed and gllamm

10.50 - 11.10 Pausa caff�

11.10 - 13.00 Sessione II - Panel Data Models, Random Coefficient and Multilevel Models

Change score, ANCOVA and Growth Curve models

Trajectory classes

Models with multiple levels

13.00 - 14.00 Pranzo

Sessione III - Generalized Linear Mixed Modelling

Generalized linear models

Subject specific and population average models

The xt suite

15.20 - 15.40 Pausa caff�

Sessione IV - Extension to Latent Variables

Introducing correlation among equations

Generalized instrumental variable model

Two-parameter item response (IRT) models

Multi-level factor models

9.00 - 13.00 - Introduction to Programming with Mata in inglese

The Introduction to Programming with Mata course will be given by Alan Riley and Bill Rising of StataCorp. This short course is an introduction to programming with Mata, Stata's fast, compiled matrix-programming language. The course will cover the basics of Mata, including types, operators, functions, and structures. The use of Mata both interactively and in batch mode from Stata do-files and ado-files will be illustrated. Mata's power and speed as a programming tool for data management and statistics will be demonstrated.

Types, operators, functions, and structures
Calling from do- and ado-files
Data and file management
Performing statistical analysis

via Rettangolo, 12/14- 67039 - Sulmona (AQ) - Italia