Variation in outcome measures

This forum is for posts that specifically focus on Ngene.

Moderators: Andrew Collins, Michiel Bliemer, johnr

Variation in outcome measures

Postby melanie » Fri Mar 05, 2021 1:49 pm

Hi,

Just a quick question, I am noticing now that sometimes when I run the exact same design a second or third time, after closing and then reopening Ngene, I get a different D-error and S-estimate (all scores are different but these are what I look at most at this stage). I am just wondering is something wrong if this happens?

Regards,
Melanie
melanie
 
Posts: 15
Joined: Thu Feb 04, 2021 7:29 am

Re: Variation in outcome measures

Postby Michiel Bliemer » Fri Mar 05, 2021 3:18 pm

What do you mean with "run the exact same design", do you mean (i) evaluate the same design in Ngene, or (ii) run the same syntax in Ngene to generate a design. If the former, that should not happen, each design has only a single D-errpr and S-estimate. If the latter, algorithms in Ngene use random search strategies to locate an efficient design. Note that are generally billion and billions of possible designs and Ngene is not able to evaluate all of them, so it simply starts with a random designs keeps improving it. Therefore, each time you run the syntax again, it will find a different design, but the D-error and S-estimates will be similar. There is not just 1 efficient design, there are many many designs with similar efficiency and Ngene locates one.

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 9 guests

cron