Optimizing detection, parameter estimation, and sky localization of gravitational wave signals using a fast marginalization algorithm

In this talk I introduce a fast algorithm for marginalizing the likelihood of compact binary merger signals over seven "extrinsic" parameters, namely the event's location, orientation and time. Since the dependence of the gravitational wave signal on these parameters is analytically known, the computation requires a single waveform query. Waveforms including higher harmonics and precession are accommodated. I show three applications of this algorithm: as a piece of the detection statistic in a search pipeline incorporating higher modes, as a tool for efficient and robust parameter estimation, and as a means of producing sky localization maps in low latency for electromagnetic follow-up of gravitational-wave alerts. An implementation is available in the parameter estimation software `cogwheel`.