X ITALIAN STATA USERS MEETING
Florence, 14-15 November 2013
Programma
9.00 - 9.25 Registration
9.25 - 9.30 Welcome from the Scientific Organisers
9.30 - 10.30 I SESSIONE – INVITED SPEAKER
The role of sensitivity analysis in estimating causal pathways from
observational data, George Ploubidis (London School of Hygiene and
Tropical Medicine)
Causal inference in observational data is a nearly alchemic task, since
parameter estimates depend on the model being correctly specified.
Researchers strive to include all potential confounders in their models, but
this assumption cannot be directly tested. Further complications arise in
causal mediation analyses where the decomposition of effects to direct and
indirect is of interest. We argue that sensitivity analysis is an effective
method for probing the plausibility of this non-refutable assumption. The
goal of sensitivity analysis in the context of causal mediation is to quantify
the degree to which the key assumption of no unmeasured confounders
must be violated for a researcher's original conclusion regarding the
decomposition to direct and indirect effects to be reversed. Three general
scenarios where the assumption of no unmeasured confounders is violated
will be discussed and results derived from sensitivity analyses appropriate
for each scenario will be presented.
10.30 – 10.50 Coffee break
10.50 - 12.30 II SESSIONE – USER WRITTEN COMMANDS AND ROUTINES I
rfmm: a Stata command for the Minimum Density Power Divergence
estimation of finite mixtures of regression models – Federico Belotti
(CEIS, Universitą degli Studi di Roma “Tor Vergata”) and Partha Deb
(Hunter College and NBER)
Estimating net survival using a life table approach – Enzo Coviello (ASL BT
Barletta), Paul W. Dickman (Karolinska Institutet), Karri Seppä (Finnish
Cancer Registry) and Arun Pokhrel (Finnish Cancer Registry)
xsmle: a Stata command for spatial panel data models estimation –
Federico Belotti (CEIS, Universitą degli Studi di Roma “Tor Vergata”),
Gordon Hughes (University of Edinburgh) and Andrea Piano Mortari (CEIS,
Universitą degli Studi di Roma “Tor Vergata”)
Reti sociali ed inizio dell’abitudine al fumo: risultati di uno studio pilota fra
gli studenti di una scuola secondaria di secondo grado – B. Federico,
G. Roscillo, S. Digennaro, A. Di Marco, G. Damiani, D. Anastasi and G.
Capelli (Universitą di Cassino e del Lazio Meridionale)
12.30 -13.15 III SESSIONE – EXPLOITING THE POTENTIAL OF STATA 13
Generalized Structural Equation Modeling in Stata – Chuck Huber
(StataCorp)
Stata's structural equation modelling (SEM) capabilities have been greatly
expanded in version 13. Support for categorical and count outcomes as
well as multilevel data structures allow us to fit a dizzying array of
models. This talk will demonstrate how to use these new features with
several common applications
13.15 – 14.30 Lunch
14.30 - 16.00 IV SESSIONE – USER WRITTEN COMMANDS AND ROUTINES II
16.00 - 16.15 Coffee break
16.15 - 17.00 V SESSIONE – EXPLOITING THE POTENTIAL OF STATA 13
Estimating average treatment effects from observational data using
–teffects- – David Drukker (StataCorp)
After reviewing the potential-outcome framework for estimating
treatment effects from observational data, this talk discusses how to
estimate the average treatment effect and the average treatment effect on
the treated by the regression-adjustment estimator, the inverse-probability
weighted estimator, two doubly robust estimators, and two matching
estimators implemented in –teffects-.
17.00– 17.30 VI SESSIONE – REPORT TO USERS – WISHES AND GRUMBLES
David Drukker and Chuck Huber (StataCorp)
20.30 Conference dinner (Optional)
CORSO DI FORMAZIONE (in italiano)
Analisi Esplorativa di Dati Spaziali
L’obiettivo del corso č quello di offrire una introduzione elementare
alla logica e alla pratica dell’analisi esplorativa di dati spaziali
mediante il software Stata. Dopo una breve presentazione delle
caratteristiche specifiche dei dati spaziali e della loro analisi, il
corso illustra – in termini sia formali che pratici – alcuni dei modi
in cui tali dati possono essere esplorati al fine di trarre utili
elementi di conoscenza sui fenomeni oggetto di studio. Tutte le
tecniche di analisi prese in considerazione saranno esemplificate
utilizzando sia i comandi standard di Stata, sia una serie di comandi
specializzati gratuitamente accessibili in rete (spmap, spgrid, spkde,
spatwmat, spatgsa, spatcorr, spatlsa). Al termine del corso i
partecipanti saranno in grado di usare Stata per visualizzare e
descrivere in termini quantitativi la configurazione dei fenomeni
spaziali di interesse.
SESSIONE I: I DATI SPAZIALI
Cosa sono i dati spaziali
Tipi di dati spaziali
La contiguitą spaziale
Matrici di ponderazione spaziale
Il comando spatwmat
SESSIONE II: LA RAPPRESENTAZIONE GRAFICA DEI FENOMENI SPAZIALI
Cartografia e GIS
Introduzione alla cartografia tematica
Il comando spmap
Stime kernel delle funzioni di densitą e di probabilitą
I comandi spgrid e spkde
Rappresentazione grafica delle funzioni di densitą e probabilitą mediante spmap
SESSIONE III: AUTOCORRELAZIONE E CLUSTER SPAZIALI
Autocorrelazione globale e locale
Indici di autocorrelazione spaziale globale e correlogramma spaziale
I comandi spatgsa e spatcorr
Indici di autocorrelazione spaziale locale e Moran scatterplot
Il comando spatlsa
Cluster spaziali
SESSIONE IV: CONCLUSIONE
Breve panoramica delle tecniche avanzate di analisi dei dati
spaziali
Potenzialitą e limiti dell’analisi spaziale
CORSO DI FORMAZIONE (in inglese)
Structural Equation Modelling in Stata
Most courses on structural equation modelling concentrate on the use of traditional
factor analytic models with interval level data and neglect the latest developments
within the Generalised Latent Variable Modelling framework. The course
will discuss the current state of the art with respect to structural
equation modelling, while retaining a practical focus. Participants
will acquire awareness of the new availablemethods and gain competence
in applying and combining these in simple settings.
MORNING SESSIONS: (Lecture and Practicals)
Latent variables , they’re everywhere: A primer on latent variable models
AFTERNOON SESSIONS: (Lecture and Practicals)
Path Analysis and SEM: An Introduction
Participants are expected to be applied statisticians, social
scientists and epidemiologists with an interest in structural equation
modelling. Familiarity with regression models would be an advantage. By
the end of the course participants should be able to:
- Understand measurement principles such as reliability and precision
- Estimate and interpret path analytic and structural models
- Develop awareness of the common threads across these methods
- Gain practical experience in using these methods using Stata
GENERAL INFORMATION:
Course participants will be provided with: a printed copy of the course handouts,
the Stata do files and databases used through the course; and a temporary licence
of Stata 13 valid for 30 days. Participants wishing to follow the empirical
applications during the course must bring their own PC.
Student prices will only be given to those participants who are able to provide
proof of their student status for the academic year 2013/2014.
Maximum number of participants permitted:
Analisi Esplorativa di Dati Spaziali – 25
5 places are currently reserved for students on this course
Structural Equation Modelling in Stata – 15
5 places are currently reserved for students on this course
|
