No Subject

[cocchi@stat.unibo.it]

Nell'ambito delle attivitą del dottorato in Metodologia statistica per la
ricerca scientifica, il 24 e 25 giugno prossimi, presso il Diaprtimento
di Scienze Statsitiche, in via Belle Arti, 41, verrą svolto un corso
breve su "Small area Estimation". Tutti  gli interessati
sono invitati a partecipare. 

cordiali saluti

daniela cocchi


Dottorato in metodologia statistica per la ricerca
scientifica, Universitą di Bologna

Introduction to Small Area Estimation


Prof. Partha Lahiri

        
Time:           

24-25
giugno 2002, ore 9,30-11 11.30-13.00

Partha Lahiri

Course Description

The course begins with a history of small-area estimation
and different uses of small-area statistics in both public and private
sectors. It will then discuss and examine various direct and indirect
small-area estimation methods available in the literature. Different
methods will be explained using simple examples. Data analyses using
several real life examples will be presented. The course is not designed
to provide an in-depth study of any topic, but to provide an overview of
small-area estimation. The course is intended for quantitatively oriented
students. Formulas will be presented wherever necessary to explain some
of the advanced topics but without any derivations.
 
Syllabus

1.      

Introduction
What is small-area?
A historical note
Issues in small-area estimation
Need for overall strategy
A few real-life applications 

2. Design-Based Estimators
      The Horvitz-Thompson Estimator
      Sampling weights
Various design-issues
 
 3. Demographic Methods
        Vital rates method
         Census component
method
        Administrative record
method
        Housing unit method
        Symptomatic regression
method
            
Ratio correlation method
            
Difference correlation method
            
Sample regression method

4. Traditional Indirect Estimation
       Synthetic estimation
        Structure preserving
estimation
        Composite Estimation
           
Sample-size dependent estimation
           
Optimal composite estimation
         A simulation study
 
5.  Model-Based Estimation
        Empirical Bayes
estimators
        Empirical Best Linear Unbiased
Predictors
        Hierarchical Bayes
Estimators
        Constrained Bayes and
Empirical Bayes Estimators

 6. Raking
        One-Way Raking
        Two-Way Raking
  
Course lecture notes and selected papers in small-area estimation and
related topics are collected in  coursepack available at the
Dipartimento di Scienze Statistiche, Universitą di Bologna (prof. Daniela
Cocchi).


Partha Lahiri is Milton Mohr Distinguished Professor of Statistics at the
University of Nebraska-Lincoln and an Adjunct Professor of the Joint
Program in Survey Methodology located at the University of Maryland at
College Park.  Professor Lahiri is also a Senior Research Scientist
of the Gallup International Research and Educational Center and an
elected member of the International Statistical Institute.  He
received a Ph.D. degree in Statistics from the University of Florida,
Gainesville, in 1986.  Formerly, Professor Lahiri was an ASA/NSF
Senior Research Fellow at the U.S. Bureau of the Census and the U.S.
Bureau of Labor Statistics.  He has been working on developing
small-area estimation methodology for various federal, state, and private
agencies for the last ten years.  His research interests include
various research topics in survey sampling, resampling methods and the
Bayes and empirical Bayes estimation.

Daniela Cocchi

Dipartimento di Scienze Statistiche "Paolo Fortunati"

Via delle Belle Arti 41

40126 Bologna

Italy

tel. +39 051 2098234    (new: compose the 0 before 51)

fax +39 051 232153    (new: compose the 0 before 51)

web page

http://www2.stat.unibo.it/cocchi/default.html