Factorial

[macartan]

1. Remove weights from estimation
2. Allow people to provide potential outcomes as formula Y = 1+ 2*T1 etc which can be transformed into outcome_means using sthg like
   If(is.null(outcome means)) outcome_means <- with (expansion.grid(k..)...., formula)

[jaspercooper]

Number 2 sounds hard. @clabima since this is your baby do you feel OK taking it on? happy to try too if you prefer.
Number 1 sounds easy though, oder? Could we break these out and do 1 for the patch and 2 for 0.2.0?