create a table about summary of 30 tables

[changxuosu]

Dear SAS community,

I have a question that needs your wisdom.

I have 30 tables, say Table1,table2,...table 30. All of them have the same data structure and the same list of variables (which includes a variable name GOOD) except that the ith  table denotes the month i.

Now I need to create a new table X, which is composed of two columns/variables.

The first variable of table X is i, the index for table i; the second variable of table X is the means of variable GOOD for table i.

Does any one can point me to any idea/procedure/SAS macro that achieve this?

Rather than

Thanks in advance!!!

[Reeza · Accepted solution]

That's one reason to never split your data sets. You can create a view which stacks the data sets temporarily and then calculates your summary statistics. Or you can write a macro that calculates it for each data set and then appends it but that's a much more complicated task. I'd only recommend that if your data sets are too big to process as one data set. 

data _temp / view=_temp;
set table1-table30 indsname=mySource;
ds = mySource;
keep Good DS;
run;

proc means data=_temp noprint;
by ds NOTSORTED;
var Good;
output out=want mean=Mean_Good;
run;

[changxuosu]

thanks, works like a charm...