Question

I have a data set with the name final_data which has numerical fields and some string fields. what i want to do is this

Print out a correlation matrix between all numeric variables in the data set and Compute the mean, min, max and number of missing for all the numeric variables in the data

Now i know how to calculate mean min and max by specifying the variables explicitly but i have no clue how to do it for numerical values. Also i dont know how to calculate number of missing values. As for correlation matrix between all numerical fields, i have no clue how to do that.

PS for column names u may use num1 num2 str1 str2 so on for numerical and string columns respectively.

Answer 1

Statistical procedures usually act on all numeric variables, so you actually don’t need to specify them, e.g.:

proc corr data=sashelp.prdsale;
run;

proc means data=sashelp.prdsale mean min max nmiss;
run;