True North Strong and…Amazing at Astrostatistics!

Figuring out how galaxies evolve over time has always been a tricky problem because we only get to see "snapshots" of different galaxy populations over time. Connecting these snapshots together into a coherent picture requires (1) inferring the evolutionary history of individual galaxies as well as (2) modelling underlying changes in populations across time using either numerical simulations or flexible statistical methods. I will highlight how advances in capturing more complex galaxy evolutionary histories and more diverse galaxy populations have substantially revised our view of how galaxies evolve over time (and improved agreement with simulations). This will cover topics including sampling methods, machine learning, and other related areas. 

The European Space Agency's Gaia mission, launched in 2013, is measuring precise positions and sky motions for about a billion stars. Most of these stars are in the Milky Way galaxy, but a small fraction are in other, nearby galaxies; an even smaller fraction are not stars at all, but solar system asteroids, distant galaxies or quasars. The enormous homogeneous database produced by the Gaia team's careful and well-documented data processing provides many opportunities for astrostatistical analysis in areas including (un)supervised classification, hierarchical Bayesian modeling, data-driven modelling, and more. This talk will review the Gaia data products and how they might drive future developments in astrostatistics, using Gaia detections of stars and other objects within nearby galaxies as one example.
 

Gravitational waves (GWs) come in many shapes and forms. Different types of signals present unique statistical challenges, requiring specialized techniques to extract a signal from noisy detector data. The GW signals observed thus far are transient, short-duration signals originating from compact binary coalescences (CBCs), i.e. mergers between black holes and neutron stars. Bayesian inference techniques have proven to be extremely powerful tools for extracting the source parameters of these merging binaries. Another class of GWs are the as-yet undetected "continuous waves" (CWs), created by rapidly rotating neutron stars. Unlike CBC signals, CWs are long-lived narrowband signals with amplitude orders of magnitude below the noise in the detector, requiring an entirely different set of search techniques. Adding to the difficulty of CW searches is the possibility that a signal may wander in complicated ways, which will need to be tracked over long periods of time. In this talk I will discuss the contrasting statistical challenges one faces when trying to study these two classes of signals, as well as the techniques employed by the GW community to tackle them. 

Fast radio bursts (FRBs) are millisecond-duration, bright, extragalactic radio flashes of unknown physical origin. Some FRB sources exhibit repeat bursts, that is, multiple bursts consistent with being emitted from the same physical source. CHIME/FRB (the FRB backend of the Canadian Hydrogen Intensity Experiment) has increased the total number of FRB detections by an order of magnitude with our teams latest catalog release of 536 FRBs. As we run our experiment for longer and our number of FRB detection grows, however, the probability of identifying two or more FRB sources within a typical localization region becomes non-negligible. A question of great importance is then, for a given repeater candidate, what is the probability that each of the bursts in the cluster are physically unrelated to one another (i.e., that they coincided by chance)? In this project, our collaborative research team is working to develop and predict an estimate of the chance coincidence probability of multiple FRBs in the case of a noisy and inhomogeneous spatial point process.