Hi everyone,
I want like to quantify a treatment effect in a model by measuring the differential effect of a treatment on a 'treatment group' versus a 'control group'. Because it is a panel data set, the diff-in-diff approach is used . Nevertheless, I would like to introduce weights to mitigate potential effects of differences across subsamples on my empirical estimation. I am using entropy balancing for the control sample to equalize the distribution of determinants across treatment and control samples.
However, I do not understand how entropy balancing is applied to the panel data set. In some papers I have read that the weights are only calculated on the basis of the pre-period, so for each unit the weights of the pre-period are used.
The question I ask myself is, for what reason uniform weights are used for each unit and they dont use different weights for each observation of a unit (the exact calculated weight for each observation)?
For example, if there are several observation rows for one company (several years), the weight of the pre-period is also used for observations of the company as a whole (also for the post-period).
Why is the weight of the pre-period used for the entire company and they don't use the specific calculated weight for each period?
Here is an example of a paper in which entropy balancing is applied in a panel data setting: http://dx.doi.org/10.2139/ssrn.3403486
In the paper they mention: Matching was performed based on the pre-reform (before 2013)
Thank you very much for your help!
0 Response to Entropy Balancing in Panel Data Setting
Post a Comment