Hello,
we are confused about the opossum package. In the blog entry it is described that the package is built on the basis of the partial linear regression model (PLR). However, the model does not return homogeneous constant values as treatment effect, even if this is selected in the generate_treatment method. Furthermore, no treatment effect is returned for unassigned observations?!
Please explain how to calculate the ATTE (average treatment effect of the treated) from this data with missing heterogenous treatment effects for the non-assigned observations.
Many thanks and best regards
Jan
Hello,
we are confused about the opossum package. In the blog entry it is described that the package is built on the basis of the partial linear regression model (PLR). However, the model does not return homogeneous constant values as treatment effect, even if this is selected in the generate_treatment method. Furthermore, no treatment effect is returned for unassigned observations?!
Please explain how to calculate the ATTE (average treatment effect of the treated) from this data with missing heterogenous treatment effects for the non-assigned observations.
Many thanks and best regards
Jan