choice-metrics.com

Dear Professor Bliemer,
I have looked at a couple of other posts on this forum about including Asc but am still a little confused.
I have an unlabelled design with 3 alternatives ( including optout).
Since there is no information about priors I assumed all priors to be zero.
My question is ,
is it correct to assume zero for the prior of the attribute-specific constant (ASC) for optout. .--> u(optout)=b6[0].
Best regards,
Sara


design
;alts= alt1*, alt2*, optout
;rows= 12
;eff=(mnl,d)
;model:
u(alt1)=b1.dummy[0|0]*color[0,1,2]
+b2.dummy[0]*newtechnique[0,1]
+b3.dummy[0]*functional[0,1]
+b4.dummy[0|0]*method[0,1,2]
+b5*price[4,7,10]/

u(alt2)=b1*color
+b2*newtechnique
+b3*functional
+b4*method
+b5*price/
u(optout)=b6[0]
$

Yes this is correct. You need to include a constant for the optout, and you can set all priors to zero (including the prior for the constant) if you have no information.

Michiel

Dear professor Bliemer,
Thank you so much for your time and reply.