Comparing exoplanet populations: Controlling for biases in host star properties by matching distributions

Abstract: Known correlations between stellar and exoplanet properties play havoc with exoplanet demographic studies that aim to compare two or more populations. For example, planet mass scales with stellar metallicity and mass. Thus, if one’s stellar subsamples have different mass or metallicity distributions, one can expect to observe differences in their giant planet occurrence rates. Controlling for such correlations is essential for robust exoplanet demographic studies. We present a novel method to control for unwanted differences between host star populations, namely, multi-variate bin matching. We apply the method to match subsamples’ mass, effective temperature, luminosity, metallicity, surface gravity, and distance distributions to investigate differences between exoplanet demographics in regions of low and high stellar position-velocity clustering. As well as studies of known exoplanets, multi-variate bin matching may be employed to achieve balanced target selection for stellar and exoplanet surveys, such as the Ariel and PLATO missions.