When to stop the Ngene iteration

This forum is for posts that specifically focus on Ngene.

Moderators: Andrew Collins, Michiel Bliemer, johnr

When to stop the Ngene iteration

Postby xiaojin » Wed Nov 27, 2019 4:27 pm

Dear Ngene expert:
I’m more thankful than I can express for your previous reply.
This is my syntax:
Design
;alts=alt1*,alt2*,alt3
;rows=24
;block=4
;eff=(mnl,d)
;model:
U(alt1)=b1[-0.0016303]*A[100,250,400,550]+b2.effects[0.1388708|0.1552699|0.0522762]*B[3,2,1,0]+b3[0.0003202]*C[1000,2000,3000,4000]+b4[0.0078205]*D[0,20,40,60]+b5.effects[0.2688591]*E[1,0]/
U(alt2)=b1*A+b2*B+b3*C+b4*D+b5*E/
U(alt3)=b0[0]
$

I'm doing an unlabeled design (efficient design). When I run Ngene with the above syntax, I find that I can't accurately judge which is the final design in “output window”. Sometimes after 15 minutes of no new iterations, I think that's the final design. But a new iteration appears in the output window half an hour later.
So I want to ask, how can I decide when to stop iterating? If I've used the final design what I thought generated the questionnaire (but actually after half an hour or 45 minutes there is a new design appears in the output window), So should I void the questionnaire I have done before?

Thank you very much!
xiaojin
 
Posts: 20
Joined: Thu Aug 29, 2019 4:55 pm

Re: When to stop the Ngene iteration

Postby Michiel Bliemer » Wed Nov 27, 2019 6:09 pm

Ngene will keep evaluating new designs unless you stop it. There are billions of possible designs and therefore it will never stop. However, at some point improvements in efficiency are marginal and any design you use should be fine. In your case, I would stop the algorithm probably after running it for perhaps one hour and then take the best design found at that moment.

Michiel
Michiel Bliemer
 
Posts: 1705
Joined: Tue Mar 31, 2009 4:13 pm


Return to Choice experiments - Ngene

Who is online

Users browsing this forum: No registered users and 12 guests

cron