Splitting in decile dataset depending on numeric variable's value for each char varaible's value

[luciacossaro]

Hi;

I have a dataset with a variable "ID" (char) and a variable "Price" (num) and some others variables.

I need split the dataset in 10 deciles (10% of observations in each) FOR EACH ID'S VALUE depending on price's values for after to can choose the highest 10% price's value for each ID'S value.

For the differents ID's values there is not the same number of observations.

I tried with PROC RANK but i get deciles for all the dataset (not depending of each ID's value).

It should be relatively easy to do it but I am a new SAS user and I can not do it.

Do you have any idea ?

Thank you!

[SuryaKiran]

Please post some sample data and the PROC RANK code you used. You might be missing by variable in PROC RANK.

[ballardw]

Try 1) sorting the data set by your ID variable

      2) add a BY Id; statement to proc rank