creating a variable number of suppliers

[DORRA SELLAMI]

Hello everyone,
I have a dataset with France as an importer and the EU countries as exporters. My dataset is at product level hs6 and covers 13 years from 2003 to 2015. Finally, I have v the value of imported product of a given year from a given country.
I want to create a variable a variable: number of suppliers for a given product i.e the number of countries from which France has imported a product in a given year.
I'd really appreciate your help for the code
Thank you very much

[Clyde Schechter]

It is much easier to give concrete advice when you give an example of your data. You have left it to me to imagine what your variable names are and how they are organized.

I'm going to assume that you have a variable called country which gives the name of the exporter, a variable called product that names or encodes the product, and another variable called year. I also assume that if in a given year a particular country did not export any of a given product to France, then there is no observation in the data set for that country-year-product combination. I also assume that you never have two entries for the same combination of exporter, product, and year: it is all summed up in a single observation.

//    VERIFY ONLY ONE OBSERVATION PER PRODUCT
//    YEAR COUNTRY COMBINATION
isid product year country, sort
by product year: gen n_suppliers = _N

If these assumptions do not adequately describe your data and you do not know how to transform your data so that they do, and you don't know how to modify the code so that it conforms to the data you have, feel free to post back. But in doing so, don't give us descriptions: show a sample of your data, and do it using the -dataex- command. Please read FAQ #12 for complete instructions on how to do that.

[Marcos Almeida]

Hello Dorra,

Welcome to the Stata Forum / Statalist.

Unfortunately, you didn't present a display of the dataset (as recommend in the FAQ), which is the best way to get a helpful advice.

That said, you may wish to fiddle with something like:

. egen myvar = total(suppliers), by(country year)

P.S: crossed with Clyde's reply.

[DORRA SELLAMI]

It is much easier to give concrete advice when you give an example of your data. You have left it to me to imagine what your variable names are and how they are organized.

I'm going to assume that you have a variable called country which gives the name of the exporter, a variable called product that names or encodes the product, and another variable called year. I also assume that if in a given year a particular country did not export any of a given product to France, then there is no observation in the data set for that country-year-product combination. I also assume that you never have two entries for the same combination of exporter, product, and year: it is all summed up in a single observation.

// VERIFY ONLY ONE OBSERVATION PER PRODUCT
// YEAR COUNTRY COMBINATION
isid product year country, sort
by product year: gen n_suppliers = _N

If these assumptions do not adequately describe your data and you do not know how to transform your data so that they do, and you don't know how to modify the code so that it conforms to the data you have, feel free to post back. But in doing so, don't give us descriptions: show a sample of your data, and do it using the -dataex- command. Please read FAQ #12 for complete instructions on how to do that.

Hello, thank you for your answer. I am a new user of Stata. I am sorry if it's not clear. This is an example of my dataset.
i is the exporter (EU country)
j is France the importer
hs6: is the product code at 6 digit level
q: is the quantity
As i said before,I want to create a variable a variable: number of suppliers for a given product i.e the number of countries from which France has imported a product in a given year

* Example generated by -dataex-. To install: ssc install dataex
clear
input int(t i j) str6 hs6 float q
2014 12 251 "100110" 8.982372
2014 12 251 "100190"   17.368
2009 12 251 "100630"       20
2013 12 251 "100630"       52
2010 12 251 "100640"        5
end

[Clyde Schechter]

So your data match what I described, except that you use different names. All you have to do is replace my variable names by the corresponding variables in your data set:

isid hs6 t i, sort
by hs6 t: gen n_suppliers = _N

[DORRA SELLAMI]

So your data match what I described, except that you use different names. All you have to do is replace my variable names by the corresponding variables in your data set:

isid hs6 t i, sort
by hs6 t: gen n_suppliers = _N

Hello Mr Clyde,
Thank you again for your answer. I run the code and it worked. Thank you very very much :D

[DORRA SELLAMI]

Hello Dorra,

Welcome to the Stata Forum / Statalist.

Unfortunately, you didn't present a display of the dataset (as recommend in the FAQ), which is the best way to get a helpful advice.

That said, you may wish to fiddle with something like:

. egen myvar = total(suppliers), by(country year)

P.S: crossed with Clyde's reply.

Hello, thank you for your answer. it worked :D thank you very much

[Marcos Almeida]

Thank you for informing your query reached it closure satisfactorily.