Dear all,
I have a question about propensity score matching for a panel data file.
The aim of my study is to understand the impact of having the first child on the wage of women. My data goes from 2001 to 2018.
To be precise:
  1. My treatment group is composed of women who gave birth in a period of time "t" which goes from 2004 to 2014; my control group is composed of women who don't have a child
  2. Women in the treatment group are treated in different years of period t, that is they gave birth in different years. For example, woman A could give birth to the first child in 2005 but woman B could give birth to her first child in 2007
Ideally what I want to do by implementing a PSM is:
  1. I want to compare women who gave birth in period t (2004 to 2014) with women who didn't give birth to the first child in period t, and I want to match them in period t-1 (that is 12 months before women gave birth).
  2. I want to compare women who gave birth in a specific year (included in period t) with a women who has no child in that specific year and I want to calculate the difference in wage of mothers and non-mothers at different points in time (one year before the birth, the year of the childbirth, and 1 to 3 years after). So ideally I want to compare woman A with a no-mother in 2005 and woman B with a no-mother in 2007.
I am not sure of how to implement this matching procedure in Stata since it is the first time that I am dealing with a treatment group where the individuals receive the treatment at different points in time. Any comments or feedback would be very much appreciated.

Thank you in advance for your time.