honestdid plot does not work!

[Ali Bahrami Sani]

Hello all,

I am trying to run an ``honestdid" command in Stata18/SE. The problem is that when I run the command, I am giving the error: "__00000A: invalid syntax in ci()". However, I can see the results text, but the plot does not appear and I get that error.

Has anyone had this problem before?!

[Jared Greathouse]

Reproducible example please.

[Ali Bahrami Sani]

Hello Jared,

I get the same error in both examples, one from the GitHub repo and the other mentioned in the help file. I even put

 ci(95)

at the end of honestdid command, but it did not work.

repo example: https://github.com/mcaceresb/stata-honestdid#honestdid

local mixtape https://raw.githubusercontent.com/Mixtape-Sessions
use `mixtape'/Advanced-DID/main/Exercises/Data/ehec_data.dta, clear
l in 1/5
* Keep years before 2016. Drop the 2016 cohort
keep if (year < 2016) & (missing(yexp2) | (yexp2 != 2015))
* Create a treatment dummy
gen byte D = (yexp2 == 2014)
gen `:type year' Dyear = cond(D, year, 2013)
* Run the TWFE spec
reghdfe dins b2013.Dyear, absorb(stfips year) cluster(stfips) noconstant
local plotopts ytitle("Estimate and 95% Conf. Int.") title("Effect on dins")
coefplot, vertical yline(0) ciopts(recast(rcap)) xlabel(,angle(45)) `plotopts'
honestdid, pre(1/5) post(7/8) mvec(0.5(0.5)2)
honestdid, coefplot cached

help file example:

use LWdata_RawData.dta, clear
mata stata(_honestExampleLWCall())
honestdid, pre(1/9) post(10/32) coefplot

The error I get in both repo and help file example:

[Frode Andre]

Your code (from the help-file) works fine in StataNow/SE 18.5.

[Ali Bahrami Sani]

I also tried on StataNow18.5/SE. I don't know why, but it doesn't work on my system. I tried many coding combinations, many ways including updating packages and software. None of them worked!

[Hemanshu Kumar]

I also got the same error. However, it has gone away after doing this:

ssc install coefplot, replace

Could you also try updating your coefplot installation?

For reference, my new version of coefplot is:

*! version 1.8.6  22feb2023  Ben Jann

[Ali Bahrami Sani]

Thank you Hemanshu.

Yes. It was the "coefplot". It works now!