A Probabilistic Approach to Classifying Supernovae Using Photometric Information

This paper presents a novel method for determining the probability that asupernova candidate belongs to a known supernova type (such as Ia, Ibc, IIL,\emph{etc.}), using its photometric information alone. It is validated withMonte Carlo, and both space- and ground- based data. We examine the applicationof the method to well-sampled as well as poorly sampled supernova light curvesand investigate to what extent the best currently available supernova modelscan be used for typing supernova candidates. Central to the method is theassumption that a supernova candidate belongs to a group of objects that can bemodeled; we therefore discuss possible ways of removing anomalous or less wellunderstood events from the sample. This method is particularly advantageous foranalyses where the purity of the supernova sample is of the essence, or forthose where it is important to know the number of the supernova candidates of acertain type (\emph{e.g.}, in supernova rate studies).Comment: Version accepted for publication in Astrophysical Journa