Hi!
I am trying to estimate if different type of events (on firm level) have an impact on the gender-gap within the firm.
I have a panel of individuals over a long period of time. I have information on the hourly wage rate and the workplace for each individual - as well as a battery of characteristics (age, experience, gender etc.)
One of my events is CEO death: So, I have data for individuals over time. Some of these individuals are subject to this event on the firm level. (Since this event occurs at different periods of time, I am interested in stacking this event at event time t = 0, and then look at the DiDiD estimate for each event year, hence -5, -4, -3, -2, 0, +1, +2, +3, +4, and +5 - so I define yearly dummies for event time +- 5 years around the event t = 0.
If I simply run a fixed effect model as;
xtset individual_identifier year
xtreg wage lead5 lead4 lead3 lead2 lead0 lag1 lag2 lag3 lag4 lag5 i.year i.age if gender==woman, fe vce(cluster firm_indentification)
estimates store G
coefplot G, keep (lead5 lead4 lead3 lead2 lead0 lag1 lag2 lag3 lag4 lag5)
then I get a DiD estimate for lead5 to lag5, where the DiD estimate reflects women who are subject to an event versus women who are not. I get the same for men. But since I want to estimate the gender gap, I am looking for the DiDiD estimates across event time. But I cannot figure out a way to specify such a model in Stata. I've tried the following:
xtset individual_identifier year
xtreg wage lead5##gender lead4##gender lead3##gender lead2##gender lead0##gender lag1##gender lag2##gender lag3##gender lag4##gender lag5##gender i.year i.age, fe vce(cluster firm_indentification)
Can anyone help me in the right direction? I have attached an example of how the data looks.
I am trying to estimate if different type of events (on firm level) have an impact on the gender-gap within the firm.
I have a panel of individuals over a long period of time. I have information on the hourly wage rate and the workplace for each individual - as well as a battery of characteristics (age, experience, gender etc.)
One of my events is CEO death: So, I have data for individuals over time. Some of these individuals are subject to this event on the firm level. (Since this event occurs at different periods of time, I am interested in stacking this event at event time t = 0, and then look at the DiDiD estimate for each event year, hence -5, -4, -3, -2, 0, +1, +2, +3, +4, and +5 - so I define yearly dummies for event time +- 5 years around the event t = 0.
If I simply run a fixed effect model as;
xtset individual_identifier year
xtreg wage lead5 lead4 lead3 lead2 lead0 lag1 lag2 lag3 lag4 lag5 i.year i.age if gender==woman, fe vce(cluster firm_indentification)
estimates store G
coefplot G, keep (lead5 lead4 lead3 lead2 lead0 lag1 lag2 lag3 lag4 lag5)
then I get a DiD estimate for lead5 to lag5, where the DiD estimate reflects women who are subject to an event versus women who are not. I get the same for men. But since I want to estimate the gender gap, I am looking for the DiDiD estimates across event time. But I cannot figure out a way to specify such a model in Stata. I've tried the following:
xtset individual_identifier year
xtreg wage lead5##gender lead4##gender lead3##gender lead2##gender lead0##gender lag1##gender lag2##gender lag3##gender lag4##gender lag5##gender i.year i.age, fe vce(cluster firm_indentification)
Can anyone help me in the right direction? I have attached an example of how the data looks.
