[Statlist] RESEARCH SEMINAR IN STATISTICS - UNIVERSITY OF GENEVA

Eva Cantoni Ev@@C@nton| @end|ng |rom un|ge@ch
Mon May 19 09:39:56 CEST 2014


RESEARCH SEMINAR IN STATISTICS - UNIVERSITY OF GENEVA


Organisers : .
E. Cantoni - E. Ronchetti - S. Sperlich - M-P. Victoria-Feser

Friday May 23rd, 2014
at 11h15
Room M 5220, Uni Mail (40, bd du Pont-d'Arve)


Ray Chambers
(National Institute for Applied Statistics Research Australia, 
University of Wollongon)

Abstract:
M-quantile models were introduced by Breckling and Chambers (1988) as a 
quantile-like generalisation of M-estimators for location. In 
particular, the M-quantile of a random variable X is a functional of its 
distribution whose value depends on the index 0 < q < 1 of the 
M-quantile as well as on the influence function underpinning the 
M-estimator. M-quantile models represent another way of modelling 
heterogeneous data, and in this context have proved useful in 
performance measurement (Kokic et al, 1997) and in small area estimation 
(Chambers and Tzavidis, 2006). In both areas of application the standard 
alternative is to characterise unexplained heterogeneity via an 
unobservable random effect. In contrast, a M-quantile model 
characterises this heterogeneity by the value of the index q that 
recovers the observed data. Much of the theory developed for M-quantile 
models assumes X has a continuous distribution, and in this context an 
M-quantile can be considered as a type of quantile. However, this 
analogy breaks down when X has a discrete distribution. In such cases it 
is popular to model unexplained heterogeneity in X by including an 
unobservable random effect in a smooth function of the expectation of X. 
In this talk I will describe recent work on extending M-quantile 
modelling to this case. In particular, I will provide applications in 
disease mapping and in poverty and unemployment estimation.


Visit the website: http://www.stat-center.unige.ch/ResSem.html

-- 
Prof. Eva Cantoni
Research Center for Statistics and
      Geneva School of Economics and Management
University of Geneva, Bd du Pont d'Arve 40, CH-1211 Genève 4
http://www.unige.ch/ses/dsec/staff/faculty/Cantoni-Eva.html




More information about the Statlist mailing list