Home » News & Blogs » Joint Survey Constraints for Cheap!
Bookmark and Share
astrobites

Joint Survey Constraints for Cheap!

20 Feb 2020, 14:00 UTC
Joint Survey Constraints for Cheap!
(200 words excerpt, click title or image to see full post)

This guest post was written by Paul Rogozenski, a PhD student at the University of Arizona. Paul studies cosmology, researching the characterization of tension within cosmological sky surveys and extensions to the standard cosmological model.
Title: Reconstructing Probability Distributions with Gaussian Processes Authors: T. McClintok, E. Rozo First Author’s Institutions: Brookhaven National Laboratory, Upton, NY; University of Arizona, Tucson, AZ. Status: Accepted to Monthly Notices of the Royal Astronomical Society [open access on arXiv] Today’s astrobite takes a detour from typical observational astronomy to talk about the statistics of large sky surveys. Surveys often go through a grueling phase of theory before observations begin, where Monte Carlo Markov Chains (MCMCs) are usually used to infer model parameters from predictive data. The MCMC does not actually simulate data, but uses Bayesian statistics and what is known a priori about a physical model of interest (e.g. the standard cosmological model) to find best-fit values and their errors from inputted data. This is done by ‘sampling’ the probability distribution, or evaluating a probability distribution at a certain point in your model. The next sample is found by proposing a small change to the current sample, evaluating the probability distribution at the proposed ...

Latest Vodcast

Latest Podcast

Advertise PTTU

NASA Picture of the Day

Astronomy Picture of the Day

astronomy_pod