Estimating and Characterizing Gravitational Wave Backgrounds in Pulsar Timing Arrays with Frequentist Statistics

Three separate PTAs have all found evidence for a common-spectrum process hinting at the imminent detection of a Gravitational Wave Background (GWB) at low gravitational-wave frequencies. Such a detection will in turn pose many new questions, most of which may be answered through spectral characterization of the GWB. Current methods for spectral analyses require time-consuming Bayesian techniques and stochastic sampling in high dimensions. Our method expands on a frequentist estimation technique called the optimal statistic.

In this talk, I will discuss how we take the current implementation of the optimal statistic and make key changes to allow for an arbitrary GWB spectral shape. This acts as a rapid proxy for a Bayesian search that may take upwards of weeks to run. This method can be used for a wide range of analyses including unmodeled single-source searches, GWB component separation, and per-frequency anisotropy searches.