Drop duplicate quarterly dates

[Nils Rydstedt]

Hi,

I am working with a dataset which consisted of monthly obeservations of two variables m1 and m3. I have now converted these montly dates to quarterly dates using

 gen tq=qofd(dofm(mdate))

I then used

 bys tq : egen m_m3 = mean(m3)

and

 bys tq : egen m_m1 = mean(m1)

to generate quarterly means of m1 and m3.

Now I am left with this

* Example generated by -dataex-. To install: ssc install dataex
clear
input long(m1 m3) float(mdate tq m_m3 m_m1)
675257 1010626 469 156 1017618.7 675961.3
673473 1013138 470 156 1017618.7 675961.3
671683 1027389 471 157 1052953.6 685409.3
686539 1061767 472 157 1052953.6 685409.3
698006 1069705 473 157 1052953.6 685409.3
689486 1063959 474 158 1081558.4   700371
698217 1088696 475 158 1081558.4   700371
713410 1092020 476 158 1081558.4   700371
733974 1117443 477 159   1113815   741235
737779 1119119 478 159   1113815   741235
751952 1104883 479 159   1113815   741235
end
format %tm mdate
format %tq tq

And the only thing left to do is to drop the duplicate quarterly variables, and of course the corresponding variables of these quarters. That is I want to drop observation 2, 3, 5, 6, 8, 9, ... , 305, 306"

I have been experimenting with

 list tq if mod(_n,2)

but I only mangages to list every 2nd, 3rd etc, observation.

[Felix Scholl]

Check out the duplicates command (see help duplicates)

This will drop all duplicate observation with regard to the variable tq:

duplicates drop tq, force

[Nils Rydstedt]

Thank you a lot!!