Abstract
Obtaining individual estimates for uncertainties in redshift-independent galaxy distance measurements can be challenging, as for each galaxy there can be many distance estimates with non-gaussian distributions, some of which may not even have a reported uncertainty. We seek to model uncertainties using a bootstrap sampling of measurements per galaxy per distance estimation method. We then create a predictive bayesian model for estimating galaxy distance uncertainties that is better than simply using a weighted standard deviation. This can be a first step toward predicting distance uncertainties for future catalog-wide analysis.
