product of county fixed effects and trend

[River Huang]

Dear All, Suppose that I have the data

* Example generated by -dataex-. To install: ssc install dataex
clear
input double county_fips float year
1 1977
1 1978
1 1979
1 1980
1 1981
1 1982
1 1983
1 1984
1 1985
1 1986
1 1987
1 1988
1 1989
1 1990
1 1991
1 1992
1 1993
1 1994
1 1995
1 1996
1 1997
1 1998
1 1999
1 2000
1 2001
1 2002
1 2003
1 2004
1 2005
1 2006
1 2007
1 2008
1 2009
1 2010
1 2011
3 1977
3 1978
3 1979
3 1980
3 1981
3 1982
3 1983
3 1984
3 1985
3 1986
3 1987
3 1988
3 1989
3 1990
3 1991
3 1992
3 1993
3 1994
3 1995
3 1996
3 1997
3 1998
3 1999
3 2000
3 2001
3 2002
3 2003
3 2004
3 2005
3 2006
3 2007
3 2008
3 2009
3 2010
3 2011
end

where county_fips denotes the county code. My question is how to simplify the following, if possible?

* Generate county-specific trends to be used in the regression analyses
bys county (year): gen trend = _n

quietly tabulate county_fips, gen(county_FE_)
foreach var of varlist county_FE* {
  gen `var'_trend = `var'*trend
  drop `var'
}
gen trend_sq = trend^2
quietly tabulate county_fips, generate(county_fe_)
foreach var of varlist county_fe* {
  gen `var'_trend_sq = `var'*trend_sq
  drop `var'
}

Any suggestion is highly appreciated.

[Jorrit Gosens]

All I can say is that I cannot imagine an analysis where the creation of all the separate variables would be of use.
Simply:

bys county (year): gen trend=_n
gen trendsq = trend^2

is probably going to be a more useful approach in most cases

[River Huang]

Hi, Jorrit,
1. Thanks for the reply. But this is not what I had in mind.
2. In fact, the code was taken from a published paper (https://onlinelibrary.wiley.com/doi/...111/ecoj.12451) in Economic Journal (a top-tier Journal in economics). The authors tried to control for the so-called county (county_fips) linear trend effect.
3. Any other suggestion would be appreciated.

[Jorrit Gosens]

Well, the study highlights the use of panel data methods to analyze a problem that has been studied with non-panel methods, but the authors do not use any of the xt suite available in Stata. Had they done that, the creation of all the separate variables would not have been needed.