search

adriennekline / psmpy

propensity score matching in python
Other
50 stars 2 forks source link

Option for PS model to transformations of variables, higher order terms or interactions between confounders? #13

Open heronils opened 1 year ago

heronils commented 1 year ago

Hi, I really love the PsmPy package. Unfortunately, for a specific dataset, PSM balance has not been achieved. Is there any option for the PS model to be modified until adequate balance is achieved, for example by including transformations of variables, higher order terms or interactions between confounders? Best, Nils