Experimental design and model selection: The example of exoplanet detection

TL;DR

This paper demonstrates how the Minimum Description Length approach can be applied to exoplanet detection, illustrating the impact of experimental design on model selection and parameter space compactness.

Contribution

It introduces a method to incorporate experimental design into MDL-based model selection, making the problem well-defined even with non-compact parameter spaces.

Findings

Experimental design influences prior and posterior distributions.

Conditioning on design can compactify parameter space.

MDL approach effectively guides exoplanet detection models.

Abstract

We apply the Minimum Description Length model selection approach to the detection of extra-solar planets, and use this example to show how specification of the experimental design affects the prior distribution on the model parameter space and hence the posterior likelihood which, in turn, determines which model is regarded as most `correct'. Our analysis shows how conditioning on the experimental design can render a non-compact parameter space effectively compact, so that the MDL model selection problem becomes well-defined.