Missing values generated by lagged var

[Jags Tanw]

Dear All,

I am using panel data for the universities for 5 years, 2012-2016. I have created a lagged variable for my dependent variable (percentage of females at professors level). The value of first year, i.e. 2012 is (will be ideally) missing in entire data. So, no value for year 2012 in lagged var. But the problem is when I am running regression with xtreg command then stata is leaving year 2012 from estimation and computing the regression coefficients for 4 year, i.e. 2013-2016. It means I am loosing info for 2012 across the data. Can somebody please advice what to do with with missing values generated by the lagged var? or is it going okay?
Sorry, I could not get much help from previous posts available at STATA forum.
Many thanks.
Jags

[Maarten Buis]

Given that your data has no information for the year 2011, we cannot know the percentage of female professors in the previous year for 2012. This is a logical consequence of using lagged variables: you loose the first observation (or if you use longer lags, the first k observations). There is nothing you can do about that.

[Jags Tanw]

Dear Maarten, thank so much for your advice so promptly. I guess then decision remains whether to use lagged var or not, as previous posts also suggest. Although, my overall results improve (in terms of significance level) by using lagged var.