Date:         Tue, 9 Mar 1999 13:54:08 -0500
Reply-To:     Kattamuri.Sarma@RESPONSEINSURANCE.COM
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         "Kattamuri S. Sarma" <Kattamuri.Sarma@RESPONSEINSURANCE.COM>
Subject:      Re: Doable regression?
Comments: To: ECON <joylin@MS2.SEEDER.NET>
Content-type: text/plain; charset=us-ascii

I suggest that first you separate the housbands' records from spouse record .
rename the wife's
   characterestics with a prefix. Then merge by household id only. Both husband
and wife belong
    to the same household (in most cases).  Then run your regression.  It should
work.
    I am assuming that you have only two types of records (husband and wife, not
son or other member).
    If not you have to separate them also .
 
    Since you are merging by household id as well as order (individual id ?),
you are seeing many
    missing values..