topic Re: Divide the table on column in SAS Enterprise Guide

Divide the table on column

PravinMishra — Thu, 26 Feb 2015 21:28:05 GMT

Hi,

I have a table called employee.Which have 10 different column or variable.I want to divide the table on the basis of column means 3 col or variable in each table.

code should handel if the col or variable number increased by 1000+.

Appreciate your suggestions.

Thanks

Re: Divide the table on column

Reeza — Thu, 26 Feb 2015 23:50:59 GMT

Your question is unclear. For starters:

What do you mean by divide?

How did you come up with 3?

10 is not equally divisible by 3, what happens to the remainder?

Please post sample data/output and/or more details.

Re: Divide the table on column

PravinMishra — Fri, 27 Feb 2015 00:19:03 GMT

you can divide the table vertically.I know it's not equally divisible. that's why I have put 3 as a number so that if what ever the column remain it ges to the last table

for ex below varidable or column are there

Last_name first_name Salary Address Hire_date Start_date End_date Account minsal maxsal

so Table1 will have

Last_name first_name Salary

Table 2

Address Hire_date Start_date

Table 3

End_date Account minsal

Table4

maxsal

let me know if I can able to explain you

Re: Divide the table on column

Haikuo — Fri, 27 Feb 2015 00:41:06 GMT

This macro should be flexible to meet your needs.

%MACRO SPLIT (

SIZE=3

,LIB=sashelp

,MEM=cars

,TARGET=cars

,LEN=200

);

%LOCAL I;

data _null_;

LENGTH VAR $ &LEN.;

do i=1 to &size until (last);

set sashelp.vcolumn (where=(LIBNAME="%upcase(&lib)" AND MEMNAME="%upcase(&mem)")) end=last;

VAR=CATX(' ', VAR, NAME);

END;

CALL SYMPUTX(CATS('NAME', _N_),VAR);

IF LAST THEN CALL SYMPUTX('N', _N_);

RUN;

DATA

%DO I=1 %TO &N;

&TARGET&I.(KEEP=&&NAME&I.)

%END;

;

SET &LIB..&MEM.;

RUN;

QUIT;

%MEND;

%SPLIT (

SIZE=4

,LIB=sashelp

,MEM=column

,TARGET=col

,LEN=100

)

Good Luck,

Haikuo

Re: Divide the table on column

Reeza — Fri, 27 Feb 2015 00:48:30 GMT

Can I ask why you want to do this? You would have to keep a KEY variable across all datasets otherwise they're not useable together anymore.

If you are separating it's usually separating into a star/snowflake schema and fact table.