I have an unbalanced panel with observations at the provincial level from 2006 to 2023. I want to estimate the effect of a treatment that is introduced from 2014 onward. My empirical strategy is based on estimating a DiD and I was hoping to use command CSDID developed by FernandoRios based on Callaway and Sant'Anna (2021) to account for multiple time periods My baseline model is the following:
I then wanted to include a dummy variable equal to one for the years 2020–2021 (Covid-19 pandemic), but the results remain unchanged.
I suspect that this might be due to an insufficient number of observations during those years.
Suggestions?
Thanks.
Code:
set seed 1 gen sample = runiform()<.9 csdid y if sample==1 , cluster(id) time(year) gvar(treat)
I suspect that this might be due to an insufficient number of observations during those years.
Code:
------------------------------------------------------------ | first_t year | 0 2014 2015 2016 2017 2018 2019 2020 Total -------+------------------------------------------------------ 2006 | 10 1 2 30 25 10 7 4 102 2007 | 11 1 2 30 25 10 7 4 103 2008 | 12 1 2 30 25 10 7 4 104 2009 | 12 1 2 30 25 10 7 4 104 2010 | 12 1 2 30 25 10 7 4 104 2011 | 12 1 2 31 26 10 7 4 107 2012 | 12 1 2 31 26 10 7 4 107 2013 | 12 1 2 31 26 10 7 4 107 2014 | 12 1 2 31 26 10 7 4 107 2015 | 12 1 2 31 26 10 7 4 107 2016 | 12 1 2 31 26 10 7 4 107 2017 | 12 1 2 31 26 10 7 4 107 2018 | 12 1 2 31 26 10 7 4 107 2019 | 12 1 2 31 26 10 7 4 107 2020 | 12 1 2 31 26 10 7 4 107 2021 | 12 1 2 31 26 10 7 4 107 2022 | 12 1 2 31 26 10 7 4 107 2023 | 12 1 2 31 26 10 7 4 107 -------+------------------------------------------------------ Total | 213 18 36 553 463 180 126 72 1,908 ------------------------------------------------------------ ---------------------------------- | first_t year | 2021 2022 Total -------+-------------------------- 2006 | 11 2 102 2007 | 11 2 103 2008 | 11 2 104 2009 | 11 2 104 2010 | 11 2 104 2011 | 12 2 107 2012 | 12 2 107 2013 | 12 2 107 2014 | 12 2 107 2015 | 12 2 107 2016 | 12 2 107 2017 | 12 2 107 2018 | 12 2 107 2019 | 12 2 107 2020 | 12 2 107 2021 | 12 2 107 2022 | 12 2 107 2023 | 12 2 107 -------+-------------------------- Total | 211 36 1,908 ----------------------------------
Thanks.