Hello,
The weighted energy distance and weighted distance covariance were recently developed in Huling & Mak (2024) and Huling et al. (2023), respectively, and it would be great if you could consider incorporating these into your existing energy statistics calculations. These have applications in causal inference for assessing the degree to which two weighted groups are similar to each other and the degree to which a treatment is associated with confounders in a weighted sample. I currently use manually calculated weighted versions of these statistics in my cobalt and WeightIt packages but would love to access the fast implementations in Rfast, especially for repeated use in selecting tuning parameters.
The weighted calculations are fairly simple and reduce to the usual calculations when all weights are unity. Thank you for considering!
Noah
Hello,
The weighted energy distance and weighted distance covariance were recently developed in Huling & Mak (2024) and Huling et al. (2023), respectively, and it would be great if you could consider incorporating these into your existing energy statistics calculations. These have applications in causal inference for assessing the degree to which two weighted groups are similar to each other and the degree to which a treatment is associated with confounders in a weighted sample. I currently use manually calculated weighted versions of these statistics in my
cobaltandWeightItpackages but would love to access the fast implementations inRfast, especially for repeated use in selecting tuning parameters.The weighted calculations are fairly simple and reduce to the usual calculations when all weights are unity. Thank you for considering!
Noah