Fourth Italian Stata Users Group meeting
Photo by Kevin Connors
|
| Dates: |
Monday and Tuesday, September 24–25, 2007 |
| Venue: |
Hotel Artemide
Via Nazionale 23
Rome, Italy |
| Days attending |
Price per person* |
| Day 1 only |
€90 |
| Day 1 + course “Meta-analisi con Stata” |
€295 |
| Day 1 + course “Modeling Count Data” |
€295 |
| Optional dinner |
TBD |
|
See Registration and accommodations for more details.
*Prices include refreshments or lunch.
|
|
Announcement and call for papers
TStat S.r.l., certified Stata distributor in Italy, announces the fourth
Italian Stata Users Group meeting, to be held in Rome on September
24–25, 2007, at the Hotel Artemide, Via Nazionale 23, Rome. The
meeting will give Stata users working in different research areas the
opportunity to exchange ideas, experiences, and information on new
applications of the software. Stata users interested in contributing to the
meeting are therefore encouraged to submit their proposals to the scientific
committee. We particularly encourage proposals based on
- developing new commands or procedures;
- using Stata in previously unpublished empirical research; and
- other applications of Stata for solving problems of general
interest, for example, data management.
Monday, September 24, 2007
The meeting will be organized in four sessions.
Presentations may be in English or Italian, depending on the speaker’s
preference. The first session is reserved for the invited speaker, Nick
Horton (Department of Statistics, Smith College, USA):
Analysis of multiple source/multiple informant data
in Stata (in English)
We describe regression-based methods for analyzing multiple-source
data arising from complex sample survey designs in Stata. We use the
term multiple-source data to encompass all cases where data are
simultaneously obtained from multiple informants, or raters (e.g.,
self-reports, family members, health care providers, administrators)
or via different/parallel instruments, indicators or methods (e.g.,
symptom rating scales, standardized diagnostic interviews, or clinical
diagnoses). This is an important problem in many social science and
medical research areas. We review regression models for analyzing
multiple source risk factors and multiple source outcomes and show
that they can be considered special cases of generalized linear
models, albeit with correlated outcomes. We show how these principled
data combination methods can be extended to handle the common survey
features of stratification, clustering, and sampling weights as well
as missing reports, and how they can be fit within Stata. The methods
are illustrated using data from health services research.
The final program will be posted in early August.
Tuesday, September 25, 2007
Two training sessions will be offered on the second day of the meeting. The
courses will run simultaneously from 9:00 AM until 5:30 PM.
Meta-analisi con Stata (in Italiano)
L’obiettivo di questo corso è quello di fornire ai partecipanti la
conoscenza sia teorica che applicata necessaria per produrre
autonomamente una revisione sistematica della letteratura scientifica
per lo studio di una ipotesi ben definita. In particolare, il corso si
soffermerà sulla parte statistica del processo di revisione
sistematica, conosciuta con il nome di meta-analisi, enfatizzando
l'interpretazione dei risultati piuttosto che la parte computazionale.
La meta-analisi rappresenta una tecnica statistica consolidata nel
settore biomedico per aggregare in maniera formale i risultati
numerici provenienti da diversi studi, ed ha visto un rapido sviluppo
in questi ultimi anni per diversi motivi: la crescita esponenziale di
articoli medici pubblicati nella letteratura; gli studi
clinici/epidemiologici spesso forniscono risultati contrastanti o non
definitivi; evitare il ripetersi di studi quando esiste ormai una
sufficiente evidenza scientifica di una determinata associazione.
Modeling count data (in English)
The Modeling count data course will be given by Prof. Joseph
Hilbe. The course will discuss the logic of modeling count response
data and provide an overview of the theory and application of Poisson
and negative binomial regression. We shall give examples of how to
select the most appropriate model and how to differentiate and deal
with apparent from real overdispersion. Models discussed are: Poisson,
negative binomial (NB-2), NB-1, heterogeneous (generalized) negative
binomial, canonical negative binomial (NB-C), NB-P (where exponent is
parameterized), zero-inflated models, zero-truncated models, negative
binomial with endogeneous stratification, hurdle models, censored
Poisson and negative binomial, panel models including fixed effect,
random effect, GEE, and mixed models
Course text:
- Hilbe, Joseph M. 2007.
- Negative Binomial Regression.
Cambridge University Press.
Click here to view the call for papers in Italian.
Click here to view the announcement in Italian.
Submission guidelines
Authors interested in presenting their work should electronically submit an
abstract before June 15, 2007, to
statausers@tstat.it. The author’s name, affiliation, and telephone
number should be included in the email. The scientific committee will make
a preliminary selection on the basis of submitted abstracts by June 30,
2007. The final version of the paper must be submitted by August 30, 2007.
Registration and accommodations
Accommodation details will be posted later. Please check our site again soon.
The meeting is being organized by TStat S.r.l.
(http://www.tstat.it),
the certified distributor of Stata in Italy.
Click here for the
registration form.
For more details, click here.
For information on registration and assistance
with accommodations, please contact Lorena Romeo at TStat:
Lorena Romeo
TStat S.r.l.
via Baden Powell, 8
67039 Sulmona AQ
Italy
Tel: +39-0864-210101
Fax: +39-0864-206014
Email: lorena@tstat.it
|
|
Organizers
|
Meetings
Stata Conference
User Group meetings
Proceedings
|
