Abstract

Statistical identification of 2XMMi sources

François-Xavier Pineau (Observatoire Astronomique de Strasbourg)

S. Derriere, L. Michel, C. Motch (Observatoire Astronomique de Strasbourg)

One of the task devoted to the Survey Science Center of the XMM-Newton satellite is the statistical identification of all serendipitous X-ray sources detected in the wide field of view of the EPIC cameras. For that purpose, we have cross-correlated the recently released 2XMMi source list with several major archival catalogues such as SDSS R6 and 2MASS. For each X-ray source, probabilities of identification with associated archival entries are computed using an original method based on a classical Bayesian approach which does not rely on Monte Carlo simulations.

Identifications in the Downes catalogue of cataclysmic variables and for AGN, galaxies and stars in the SDSS R6 form the basis of our learning sample. The multi-wavelength parameter space has been reduced by a principal component analysis which takes into account measurement errors on all data. Both a knn and a more elaborated kernel density smoothing approach have been tested for the supervised classification.

We will report on the current status of this project and will show some illustrative results arising from the classification of the 2XMMi identifications in the SDSS DR6 catalogue.

Mode of presentation: poster