The Milky Way Galaxy has many natural satellites, such as globular clusters and dwarf galaxies, that can be used to estimate the mass of the Milky Way. The observed positions and velocities of these satellites may be used to constrain model parameters for the gravitational potential, and thus for the mass profile. One issue with this method, however, is that the velocity data of many satellites are incomplete--- that is, we only have their line-of-sight velocity measurements. Therefore, the scientist must either 1) rely on mass estimators that use only line-of-sight velocities (and throw out the proper motion measurements that we do have), or 2) employ a mass estimator that uses only complete data (and then toss out the data that is incomplete). Either way, the data are not being fully utilized.
We have developed a Bayesian method that remedies this issue, and allows both complete and incomplete data to be used simultaneously in the analysis. In my talk, I will describe the method and its application to the Milky Way, including our preliminary results. I will also explain the next steps we are taking to improve the analysis: including the measurement uncertainties via a hierarchical Bayesian model.
