On the likelihood-ratio test applied in asteroseismology for mode identification

Abstract

The identification of the solar-like oscillation modes, as measured by asteroseismology, is a necessary requirement in order to infer the physical properties of the interior of the stars. Difficulties occur when a large number of modes of oscillations with a low signal-to-noise ratio are observed. In those cases, it is of common use to apply a likelihood-ratio test to discriminate between the possible scenarios. We present here a statistical analysis of the likelihood-ratio test and discuss its accuracy to identify the correct modes. We use the AsteroFLAG artificial stars, simulated over a range of magnitude, inclination angle, and rotation rate. We show that the likelihood-ratio test is appropriate up to a certain magnitude (signal-to-noise ratio).