Hello
I am working with csdid command and I have an inflation variable. So, I am trying to assess the impact of a policy by looking at inflation, which is the only variable I have. Now, I am trying to find out a threshold level of inflation, beyond which the policy will not work or I would not get any significant results from my estimation. I created a loop for this but decided to drop the outlier observations. I have country-level data but instead of dropping outlier countries I only dropped the observations. Hence, even though I started with a balanced panel, I have an unbalanced panel after dropping outlier observations.
How important is it to maintain a balanced panel for csdid command to work? Also, I thought that instead of dropping, I could control for those observations, but I am sure how to include that in my code.
Thank you!
I am working with csdid command and I have an inflation variable. So, I am trying to assess the impact of a policy by looking at inflation, which is the only variable I have. Now, I am trying to find out a threshold level of inflation, beyond which the policy will not work or I would not get any significant results from my estimation. I created a loop for this but decided to drop the outlier observations. I have country-level data but instead of dropping outlier countries I only dropped the observations. Hence, even though I started with a balanced panel, I have an unbalanced panel after dropping outlier observations.
How important is it to maintain a balanced panel for csdid command to work? Also, I thought that instead of dropping, I could control for those observations, but I am sure how to include that in my code.
Thank you!