Astrostatistics Research Area
Astrostatistics is the study of stars, galaxies, and the large scale structure of the Universe. In the last quarter century, astronomy has gone from being a data-poor discipline (catalogs with hundreds or thousands of observations) to one that is rich in data (catalogs with hundreds of millions or even billions of observations). In the late 1990s, Chris Genovese and Larry Wasserman helped found what was then called the Pittsburgh Computational Astrostatistics (PiCA) collaboration, along with astronomers and computer scientists at both CMU and Pitt. Currently, several faculty and graduate students are members of the Astrostatistics group; and other active members are drawn from the McWilliams Center for Cosmology at Carnegie Mellon as well as the Department of Physics and Astronomy at the University of Pittsburgh. Our work mainly focuses on making statistical inferences given complex high-dimensional data; for instance, estimating the redshifts (distance-analogues) of galaxies given low-resolution spectra of non-representative labeled data, studying how galaxies evolve given high-dimensional galaxy images, and inferring the spatial distribution of neutral hydrogen backlit by distant quasars.
