PROC REG: adding 2 Dummy Variables

[RazzleBayker]

Hi guys,

I have a muiltiple regression model with 2 dummy variables.  The qualitative variables in my data looks like this:

Broad	Sector
1	0
1	0
0	0
0	1
0	0

Each observation can belong to either the Broad category, Sector category, or neither.  This would require me to have 2 dummy variables.  How can I add these dummies to my model which looks like this so far:

PROC REG data=GH3 outest=est;

by symbol;

model ADV = Volume Volatility MarketCap;

(need to add my dummies to this model)

run;

quit;

data FINAL;
set est;
format Volume Volatility MarketCap e15.;
run;

proc print data = FINAL;
run;
quit;

-------------

Thanks in advance,

Razzle

[DBailey · Accepted solution]

I think that most of the dummy variables are added to the input dataset.  If you don't want to recreate the data, you can create a view which has the dummy variables in it.

proc sql;

create view work.gh3_view as

select

t1.*

,case when t1.category='Broad' then 1 else 0 end as Broad

,case when t1.category='Sector' then 1 else 0 end as Sector

from work.gh3;

quit;

and then use that as the input data to your proc.

[RazzleBayker]

Thanks a lot.