A cautionary tale in fitting galaxy rotation curves with Bayesian techniques

The application of Bayesian techniques to astronomical data is generally non-trivial because the fitting parameters can be strongly degenerated and formal uncertainties are themselves uncertain. An example provided by contradictory claims over presence or absence a universal acceleration scale ( g † ) in galaxies based on fits rotation curves. To illustrate this we present an analysis which Newtonian gravitational constant G N allowed vary from galaxy when curves SPARC database, analogy recently debated analyses. When imposing flat priors , obtain wide distribution which, taken at face value, would rule out as with high statistical confidence. However, empirically motivated log-normal prior returns virtually no sacrifice fit quality. This implies that inference variable (or result combined effect parameter degeneracies unavoidable error model. these effects into account, consistent (and ).