4-year interval and retardation variable

[Dongni Wang]

Dear all:

I use fixed-effect threshold model to study the effect of aid on FDI. The data is yearly with 1 year delay for control variables and variable of interest to avoid posible endogeneity problem.
Recently, I've read some relevant studies in which the authors use 3- or 5-year interval to estimate the effect. So I would like to test with my sample.
The original sample covers from 2003 to 2016 and I remove the data for 2003 and 2004, thus it can be divided into 4-year interval. 2005-2008(t1), 2009-2012(t2) and 2013-2016(t3).
My question is that by applying 4-year interval, the endogeneity problem will still existe? And if it exists, the new retarded variable now refers to t-1 not year-1? That is to say, I have generate retarded variable by: gen Xr4=l4.X ?
Thanks

Dongni

[Phil Bromiley]

You didn't get a quick answer. Such general questions sometimes don't spark interest here - other venues might be better. By the way, say lagged not retarded.

If the endogeneity really exists when you use data from t on both sides of the equation, when you average t to t+5 on both sides of the equation it will still exist.

I'm not fond of such averaging - it creates somewhat arbitrary categorizations. Instead, what about using an average of t+1 to t+5 (or whatever time period) for the dv, and still use annual observations on the rhs? You'll need robust standard errors that account for serial correlation.

Note that Stata is quite happy with L4.x as a rhs variable - you don't need to create the variables before the regression. If you want an average, you'll need to create it before the regression.