Dear Stata users,
I am running difference-in-differences (didregress) and I'm facing a problem. The interaction term (treatment year and treatment group) is omitted because of collinearity.
The dependent variable is a binary variable indicating weather or not firms perceive corruption as a problem.
The goal is to analyse the impact of demonitisation (took place in 2016) on corruption in India
The treatment group is India and the control group is Bangladesh.
The main purpose of running didregress is to check for postestimation through ptrends (Parallel-trends test) and plot the same.
Please find below the command line from stata.
Please, any help as to how to deal with this?
Thank you in advance.
Rhea.
I am running difference-in-differences (didregress) and I'm facing a problem. The interaction term (treatment year and treatment group) is omitted because of collinearity.
The dependent variable is a binary variable indicating weather or not firms perceive corruption as a problem.
The goal is to analyse the impact of demonitisation (took place in 2016) on corruption in India
The treatment group is India and the control group is Bangladesh.
The main purpose of running didregress is to check for postestimation through ptrends (Parallel-trends test) and plot the same.
Please find below the command line from stata.
Code:
gen treat_dummy = 1 if country=="India"
replace treat_dummy = 0 if country=="Bangladesh"
gen year_dummy = 1 if year>2016
replace year_dummy = 0 if year_dummy==.
gen inter = treat_dummy * year_dummy
didregress (finance) (inter), group(treat_dummy) time(year)
note: inter omitted because of collinearity.
model is not identified
The treatment variable inter was omitted because of collinearity.
Please, any help as to how to deal with this?
Thank you in advance.
Rhea.
