Rank countries in order of average GDP

[Sigrid Morrison]

Hi, I want to do something very very simple and have tried searching statalist for the command that will produce it. I have generated the average GDP for each country and would now like to list the countries in order of highest GDP to the least. I think all I need is one command after this:

egen meanGDP = mean(GDP), by (country)

Any advice would be appreciated. Thanks in advance.

[Jeph Herrin]

You say "average GDP", does this mean you have several different values (eg, years) for each country? Please describe your data.

[Roman Mostazir]

I think this should work. Assuming you have the mean for countries (meanGDP).

gsort -meanGDP /*the minus sign (-) before meanGDP sorts the variable in decending order*/
list country

[Sigrid Morrison]

Roman, thank you!
Jeph, I have several years of data, and have used the egen command to calculate one average of those years. However I've just discovered that it then applies to the country over and over, so when I list, I get

1. Luxembourg
2. Luxembourg
3. Luxembourg

I would like each country to just appear once. Thank you both for your replies.

[Roman Mostazir]

Please note Jeph's suggestion. If you have countries repeated in the dataset, the list command will produce repeated data for the countries. Best for you is to provide a sub sample of your data.

[Sigrid Morrison]

This is a small sub-subsample of the layout of the data. I would like to just make a simple table with one value for each country, ranked in descending order. I could do it in excel but don't know how to do it in STATA

country	s006	continent_n	meanGDP
Albania	200656	europe	3.437574
Albania	152	europe	3.437574
Albania	1190	europe	3.437574
Albania	534	europe	3.437574
Albania	553	europe	3.437574
Albania	524	europe	3.437574
Albania	1041	europe	3.437574
Albania	1091	europe	3.437574

[Roman Mostazir]

You need to creat a variable that identifies country just once and individually.

gsort -meanGDP
egen tag_country=tag(country)
list country if tag_country==1

[ben earnhart]

or, since meanGDP is the same for each observation within country:

bysort country: keep if _n==1

drops potentially useful cases, but can either preserve/restore, or just reload original data.

[ben earnhart]

better yet, no cases lost:

tabulate country, summarize(GDP)

[Sigrid Morrison]

Thanks so much for your help! That's the first step at least ... I know there are plenty more things to confuse me yet