How to use hp filter on different groups

[John Zhao]

Hi,

I have GDP data for each state from the last 10 years, and I want to use hp filter on each state's time series, suppose "state" is the variable for state name, and "year" is the time, is there a concise way to use "tsfilter hp" command to detrend each state's time series?

Thanks!

[Justin Niakamal]

From what you've stated, it sounds like you have your data in a long format. You can either reshape your data or create a tempfile and loop through each state.

For example (something along these lines, I can't verify w/o seeing your data)

tempfile temp

egen id = group(state)
local end = id[_N]
save `temp'
forvalues x = 1/`end' {
     use `temp' , clear
     keep if id == `x'
     tsset year
     tsfilter hp state_hp = state_gdp
     tempfile _`x'
     save `_`x''
}
clear
forvalues x = 1/`end' {
      append using `_`x''
}

[Clyde Schechter]

Disclaimer: I don't use and know little about -tsfilter-, so what I write here is predicated on the correctness of what Justin Blasongame wrote in #2. Assuming that is right, the code can be considerably simplified:

capture program drop one_state
program define one_state
    tsset year
    tsfilter hp state_hp = stage_gdp
    exit
end

runby one_state, by(state)

-runby- is written by Robert Picard and me, and is available from SSC.

[John Zhao]

Yes, you are right, it's long format, and I found that by using

encode stabr, gen(state)
tsset state year
tsfilter hp stateGDP_hp=stateGDP

the hp filter works autimatically by state and year, please correct me if I'm wrong.