Hi everyone, I have unbalanced panel data, which is required to perform a DID analysis. The dataset is about firms' use of debt and state tax changes from 1989 to 2011. My question is how to generate the post_treatment dummy variable since different states apply tax changes at different years. I already have the dummy to indicate when tax changes for different state. I have tried the following code but it only returns 1 for the year following a tax change and then back to 0 again. bysort DCompanyName ( DataYearFiscal ): generate POSTIncrease = ( HQTaxIncrease [_n-1]==1) (HQTaxIncrease is the dummy to indicate that the state apply tax increase at a specific year)
One more thing, some states might apply 2 or 3 tax changes. For example, company A survives through the whole time period, but the state it stays in changes the tax rate at 1991, 2000, and 2008. In that case, how to determine when is the post-treatment period?
One more thing, some states might apply 2 or 3 tax changes. For example, company A survives through the whole time period, but the state it stays in changes the tax rate at 1991, 2000, and 2008. In that case, how to determine when is the post-treatment period?
Comment