Date:
Thursday, August 7, 2014, 11:00am to 12:00pm
Location:
Pratt conference room (G04), CfA
Approximate Bayesian Computing in Astronomy
A standard Bayesian statistical analysis relies on the specification of a likelihood function. Unfortunately the likelihood is not always known or tractable. Approximate Bayesian computation (ABC) provides a framework for performing inference in cases where the likelihood is not available, but it is possible (and computationally efficient) to generate a sample from the forward process that mimics the data-generation process.
I will introduce and discuss ABC with a goal of illustrating how it can be useful in astronomy. Throughout, astronomical examples will be used to clarify concepts, and I will conclude with an application to exoplanet eccentricity distributions.
(Jessi Cisewski is a visiting assistant professor in the statistics department at Carnegie Mellon; her research is primarily focused on astrostatistics. Her current work is focused on inferring the exoplanet eccentricity distribution from Kepler data; she has previously worked in mapping the IGM using the Lyman alpha forest.)