Dear all,

I am working with a dataset of firms in Brazil and I need to create some descriptive statistics at the municipal level.

In particular, I would like to create two variables.

The first tells me which municipality (I have a string variable for municipality) has the largest value for a given numeric variable (this variable is named 'order_var') and the second variable tells me which municipality has the second-largest value for the same numeric variable ('order_var').

The dataset is like

municipality order_var
SALVADOR 21.5
BELO HORIZONTE 15.8
RIO DE JANEIRO 31
CANOAS 29.5
CAXIAS DO SUL 25.4

and I would like it be as the following

municipality order_var max1_var max2_var
SALVADOR 21.5 RIO DE JANEIRO CANOAS
BELO HORIZONTE 15.8 RIO DE JANEIRO CANOAS
RIO DE JANEIRO 31 RIO DE JANEIRO CANOAS
CANOAS 29.5 RIO DE JANEIRO CANOAS
CAXIAS DO SUL 25.4 RIO DE JANEIRO CANOAS

where 'max1_var' is the first variable aforementioned and 'max2_var' is the second one, respectively.

Can you help me with that?

Thank you very much.

Any help is greatly appreciated.


Obs: here is the code to import the fictitious data (Let's suppose that we have 10 firms

clear
input str25 municipality order_var
"SALVADOR" 21.5
"BELO HORIZONTE" 15.8
"RIO DE JANEIRO" 31
"CANOAS" 29.5
"CAXIAS DO SUL" 25.4
end