Spatial survival analysis (Adrian Baddeley)

[Giacomo Aletti <giacomo.aletti@unimi.it>]

          SEMINARIO DI MATEMATICA APPLICATA

Lunedi` 4 Febbraio 2002 alle ore 15.00, 
nella Sala Rappresentanza del Dipartimento di Matematica 
(Via Saldini 50, Milano),

il Prof. Adrian Baddeley, della University of Western Australia,

terra` un seminario dal titolo:

"Spatial survival analysis"

Abstract.
In spatial statistics and image analysis, one of the successful
traditional strategies for data analysis is to measure spatial
distances, such as interpoint distances in a point pattern, or the
diameters of regions in an image.

There is an analogy between distances in spatial processes and
waiting times in survival analysis. This allows us to apply the
methods of survival analysis to problems in spatial statistics.
Edge effects in spatial statistics correspond to the censoring of
waiting times in survival analysis.  Analogues of the Kaplan-Meier
estimator have been constructed.  Statistics for spatial point
patterns can be re-interpreted as products of survival analysis,
for example, the $J$-function is a ratio of survival
probabilities.

Semiparametric models in survival analysis, such as the Cox
proportional hazards model, allow us to extend this hybrid
methodology to the case of nonstationary spatial processes, and
indeed to include dependence on spatial covariates. This suggests
practical techniques for model-based analysis of spatial and image
data, even when these exhibit spatially-varying texture.

La S.V. e` cordialmente invitata.
-----------------------------------------------------------------------
Home page del seminario: http://www.mat.unimi.it/~pavarino/sma/
Il seminario e` organizzato da:
- Dipartimento di Matematica dell'Universita` di Milano 
- MIRIAM (MIlan Research group in Industrial and Applied Mathematics) 
- Dottorato di Ricerca MSSCI (Matematica, Statistica, Scienze 
Computazionali e Informatica).
Per ulteriori informazioni contattare Luca Pavarino o Kevin Payne.