Missing results in computing growth rate

[Paris Rira]

Dear Profs and Colleagues,

I am going to compute the average growth rate of one variable over the period 2010-2020. (I mean the growth rate, according to 2 years).

* Example generated by -dataex-. For more info, type help dataex
clear
input double(year_2010 year_2020 growth) str6 Region
.099 .217  .073 "Lisboa"
.078 .077 -.012 "Centro"
.057 .076   .02 "North" 
end

When I apply this code, missing values are generated. 

gen g_var = D.vari1 / L.vari1
(6 missing values generated)

*Variables are in % form. The answer is in the third column, Red values, though I have no idea where they came from.*

Any ideas are appreciated.

[Hemanshu Kumar]

Your data extract has no variable var1, so your code cannot be replicated on it. Also, you show 3 values, but your warning is about 6 missing values being generated. Has your data been reshaped after running the line of code you show?

Please show a data extract from the data as it is at the point that your code is applied to it.

[Paris Rira]

To set Panel I had to change the variablesl shape.

* Example generated by -dataex-. For more info, type help dataex
clear
input int year double vari1
2010 .057
2020 .076
2020 .077
2010 .078
2010 .099
2020 .217
end


.  tsset vari1 year

Panel variable: vari1 (weakly balanced)
 Time variable: year, 2010 to 2020
         Delta: 1 unit

. gen g_var = D.vari1 / L.vari1
(6 missing values generated)

[Hemanshu Kumar]

This data example also doesn't work, since you are missing the Region variable, which is needed for the panel. Here is some alternative code to point you in the right direction:

clear
input int year double vari1 str6 Region
2010 .057 "North"
2020 .076 "North"
2020 .077 "Centro"
2010 .078 "Centro"
2010 .099 "Lisboa"
2020 .217 "Lisboa"
end

encode Region, gen(region) // to enable tsset, which needs a numeric panel variable

tsset region year, delta(10)
gen g_var = D.vari1 / L.vari1

this produces:

. li year region vari1 g_var, noobs sepby(region)
  +-----------------------------------+
  | year   region   vari1       g_var |
  |-----------------------------------|
  | 2010   Centro    .078           . |
  | 2020   Centro    .077   -.0128205 |
  |-----------------------------------|
  | 2010   Lisboa    .099           . |
  | 2020   Lisboa    .217    1.191919 |
  |-----------------------------------|
  | 2010    North    .057           . |
  | 2020    North    .076    .3333333 |
  +-----------------------------------+

[Paris Rira]

Dear Kumar,

Thank you for getting back to me.
I agree with you, though if you look at #1 you can see that the results (Red values) do not look like yours.

[Hemanshu Kumar]

The red values for Lisboa and North make approximate sense as annualised growth rates, but that logic does not work at all for Centro. I'm afraid I cannot tell you how those numbers were computed.