Concatenating Temporary Matrices

[Jared Greathouse]

When we run the code

clear *
net from "https://raw.githubusercontent.com/jgreathouse9/FDIDTutorial/main"
net install fdid, replace


cls

u "https://github.com/jgreathouse9/FDIDTutorial/raw/main/smoking.dta"

qui fdid cigsale, tr(treated) unitnames(state)

loc T = e(T)

mkf eventframe

cwf eventframe

svmat e(didframe), names(col)

xtset id year

cap qui xtdidregress (cigsale) (treat), ///
group(id) time(year) ///
vce(bootstrap, reps(50) seed(1466) dots(10))
cls
cap qui estat grangerplot, post nodraw
ereturn list

tempname cis effs unc point

mat `cis'= e(ci_normal)'

matrix `unc' = `cis'[(rowsof(`cis')+1) - `T'..., 1..2]
cls
mat `effs' = e(b)'

matrix `point' = `effs'[(rowsof(`effs')+1) - `T'..., 1]

mat l `unc'

mat l `point'

mat effs = `point'\`unc'

we get

. mat l `unc'

__000002[31,2]
                 ll          ul
_lead19  -8.4543431   15.704343
_lead18   -11.51243   12.912432
_lead17   -17.82874   15.878735
_lead16  -19.797317   18.297321
_lead15  -17.689872   19.189864
_lead14  -17.309823   20.459825
_lead13  -15.092363   18.142366
_lead12  -9.9613119   13.211312
_lead11  -7.2719302    13.02193
_lead10  -9.3121778   10.862177
 _lead9  -8.8646666   8.9146621
 _lead8  -10.672925   10.272923
 _lead7  -11.774687   10.224688
 _lead6  -9.4302961   8.8803045
 _lead5  -12.886588   17.036593
 _lead4  -5.6067273    8.356735
 _lead3  -4.9207772   8.9207657
 _lead2   .38994673   3.2100556
  _lag0  -5.2844066  -2.6655865
  _lag1  -8.3759589  -2.4740358
  _lag2  -19.509309   3.2593053
  _lag3  -22.172043   6.9720417
  _lag4  -25.949461   5.8494668
  _lag5  -32.395063    4.045065
  _lag6  -34.984833   3.7348332
  _lag7  -33.874241   4.1742399
  _lag8  -35.182563    5.932559
  _lag9  -35.370716  -.12928449
 _lag10  -35.605481   -5.744518
 _lag11  -35.784331  -3.5656713
  _cons    104.8436    161.1064

.
. mat l `point'

__000003[31,1]
                 y1
_lead19       3.625
_lead18   .70000076
_lead17  -.97500229
_lead16  -.74999809
_lead15   .74999619
_lead14   1.5750008
_lead13   1.5250015
_lead12       1.625
_lead11       2.875
_lead10   .77499962
 _lead9   .02499771
 _lead8  -.20000076
 _lead7  -.77499962
 _lead6   -.2749958
 _lead5   2.0750027
 _lead4   1.3750038
 _lead3   1.9999943
 _lead2   1.8000011
  _lag0  -3.9749966
  _lag1  -5.4249973
  _lag2  -8.1250019
  _lag3  -7.6000004
  _lag4  -10.049997
  _lag5  -14.174999
  _lag6     -15.625
  _lag7      -14.85
  _lag8  -14.625002
  _lag9      -17.75
 _lag10  -20.674999
 _lag11  -19.675001
  _cons     132.975

I wish to concatenate these together into a one matrix such that we have our point estimates and upper and lower bounds alongside each other. But when I do this, I get

mat effs = `point'\`unc'
__000003`unc not found
r(111);

Why might this be? And how would I navigate this issue?

[George Ford]

use , not \

[Bjarte Aagnes]

Why? Because the backslash is an escape character and prevents macro substitution.

tempname a b

matrix `a' = 1,2
matrix `b' = 3,4
mat c = `a'\\`b'

matlist c