alternative weight scaling

[gravesti]

Currently we have scaled and unscaled weights. Could we scale the weights so that they sum to an arbitrary value depending on treatment group?

We are mostly interested in unanchored case?
Could this be an option in estimate_weights()?

[MikeJSeo]

I was reading over Conor's paper and another possible scaling would be to scale to ESS. Thoughts?