Labmask command with parallel trends

[Anoush Khachatryan]

Hello Members,

Below is a code taken from this post: https://www.statalist.org/forums/for...nel-data-graph

Is there a way to modify this code to include parallel trends?

I would appreciate any help!

Thanks,
Anoush

 // open some example data
clear all
use "https://www.rug.nl/ggdc/docs/pwt90.dta", clear
gen gdppc = rgdpe/ pop
eep country year gdppc pl_con  

// keep the first 35 countries
bys country (year) : gen mark =  _n == 1
replace mark = sum(mark)
keep if mark <= 35 drop mark  
// sort countries by their average price level
bys country : egen mean = mean(pl_con)
egen order = rank(mean), track
labmask order, values(country)  
// add background lines
tempfile temp
save `temp'  

separate pl_con, by(country) veryshortlabel

forvalues i = 1/35 {    
bys year (pl_con`i') : replace pl_con`i' = pl_con`i'[1]    
by year : gen gdppc`i' = gdppc[1]
}

keep if country == "Albania"
keep year pl_con? pl_con?? gdppc? gdppc??
merge 1:m year using `temp'  

local gr ""
forvalues i = 1/35 {    
local gr `gr' line pl_con`i' gdppc`i', lpattern(solid) lcolor(gs12) ||
}  

// final graph
sort order year
twoway `gr' ///    
line pl_con gdppc , by(order, legend(off) compact note("")) ///    
lpattern(solid) lwidth(*1.2) ///    
xscale(log) xlab(1000 "10{sup:3}" 10000 "10{sup:4}" 100000 "10{sup:5}") ///    
xtitle("GDP per capita (log scale)") ytitle("price level (USA 2011 = 1)")

[Jared Greathouse]

Thanks so much for including a way to get your real data- I think people should do this more often if the can. Anyways, I've corrected a few errors in your code

// open some example data
clear all
use "https://www.rug.nl/ggdc/docs/pwt90.dta", clear
gen gdppc = rgdpe/ pop
keep country year gdppc pl_con  

// keep the first 35 countries
bys country (year) : gen mark =  _n == 1
replace mark = sum(mark)
keep if mark <= 35

drop mark  
// sort countries by their average price level
bys country : egen mean = mean(pl_con)
egen order = rank(mean), track
labmask order, values(country)  
// add background lines
tempfile temp
save `temp'  

separate pl_con, by(country) veryshortlabel

forvalues i = 1/35 {    
bys year (pl_con`i') : replace pl_con`i' = pl_con`i'[1]    
by year : gen gdppc`i' = gdppc[1]
}

keep if country == "Albania"
keep year pl_con? pl_con?? gdppc? gdppc??
merge 1:m year using `temp'  

local gr ""
forvalues i = 1/35 {    
local gr `gr' line pl_con`i' gdppc`i', lpattern(solid) lcolor(gs12) ||
}  

// final graph
sort order year
twoway `gr' ///    
line pl_con gdppc , by(order, legend(off) compact note("")) ///    
lpattern(solid) lwidth(*1.2) ///    
xscale(log) xlab(1000 "10{sup:3}" 10000 "10{sup:4}" 100000 "10{sup:5}") ///    
xtitle("GDP per capita (log scale)") ytitle("price level (USA 2011 = 1)")

My only question is what do you mean by

modify this code to include parallel trends

and are you sure this, uhh, massive graph is the best medium to convey whatever you mean to convey?

[Nick Cox]

I am the author of labmask (which can be found on SSC, but I ask that people mention Stata Journal as the source).

I can't see that its use has anything in particular to do with parallel trends.

[Anoush Khachatryan]

Thank you Jared Greathouse and Nick Cox for your responses.

Using the graphs produced from the above code (shown here: https://www.statalist.org/forums/for...nel-data-graph), I would like to include a second line, per graph, for a comparison group. I am interested in the general trends of the treatment and comparison groups. If labmask isn't the best way to go about this, is there something else you would recommend?

Thanks,
Anoush

[Nick Cox]

I didn't say that labmask wasn't useful. My point was rather that I didn't see that you need anything different from it just because you are interested in parallel trends.

[Jared Greathouse]

I share Nick's confusion about the relevance of labmask here. I only ever use it when I want to label an already existing variable if I can't use egen to do so.

When I think of a comparison group, I think of plotting the average of a control group against the observed outcomes of a treated unit, or I think of a counterfactual that I've created by weighting or some other means. Are you doing a difference-in-differences analysis? What exactly are we working with here?

The trouble here is that I don't know what parallel trends has to do with the labmask command. There's not even a treatment variable/indicator here, so I'm a little confused on how labmask is relevant at all for this problem. Are you really just asking how to plot the values of one country versus the averages of another set of countries?

[Anoush Khachatryan]

I apologize for the confusion.

Are you really just asking how to plot the values of one country versus the averages of another set of countries?

Yes! That is exactly what I am asking. I am not sure how to plot this.