Abstract

Future surveys such as the Dark Energy Survey and the Large Synoptic Survey Telescope will produce an unprecedented amount of photometric supernova data, not all of which can be followed up spectroscopically. Light curve fitting techniques can provide a probability that an object is a Type Ia supernova, but contamination from other types of supernovae can lead to biases to the estimation of cosmological parameters. BEAMS (Bayesian Estimation Applied to Multiple Species) is a fully Bayesian analysis technique designed to take contamination into account and produce unbiased estimates of the parameters. BEAMS is a general technique which should be applied in any situation where contamination from other types of objects is possible. In this colloquium, I will explain how BEAMS works and how it is applied to supernova cosmology. I will also discuss my current work on extending BEAMS to deal with correlated data and comparing BEAMS with an alternative Monte Carlo approach.