QUANTIFYING QUASAR VARIABILITY AS PART OF A GENERAL APPROACH TO CLASSIFYING CONTINUOUSLY VARYING SOURCES

Abstract

Robust fast methods to classify variable light curves in large sky surveys\nare becoming increasingly important. While it is relatively straightforward to\nidentify common periodic stars and particular transient events (supernovae,\nnovae, microlensing), there is no equivalent for non-periodic continuously\nvarying sources (quasars, aperiodic stellar variability). In this paper we\npresent a fast method for modeling and classifying such sources. We demonstrate\nthe method using ~ 86,000 variable sources from the OGLE-II survey of the LMC\nand ~ 2,700 mid-IR selected quasar candidates from the OGLE-III survey of the\nLMC and SMC. We discuss the location of common variability classes in the\nparameter space of the model. In particular we show that quasars occupy a\ndistinct region of variability space, providing a simple quantitative approach\nto the variability selection of quasars.\n