Interpreting efficiency measures for efficient designs

This forum is for posts covering broader stated choice experimental design issues.

Moderators: Andrew Collins, Michiel Bliemer, johnr

Post Reply
tn73
Posts: 12
Joined: Thu May 16, 2013 9:29 pm

Interpreting efficiency measures for efficient designs

Post by tn73 »

Dear all,
When I run NGENE, one of the MNL efficiency measures listed in the output is the T estimate (and Tp estimate) and also N estimate & N star. I can’t find where these are defined or discussed in the manual. What do they measure and how can I interpret them?

Regards
Tom
Michiel Bliemer
Posts: 2055
Joined: Tue Mar 31, 2009 4:13 pm

Re: Interpreting efficiency measures for efficient designs

Post by Michiel Bliemer »

Hi Tom,

What version of Ngene are you using? You can find this under the menu Help | About.
T and N estimates are not reported in Ngene versions 1.1.x and 1.2.x, but maybe you are using an older version or special research version.

Michiel
johnr
Posts: 172
Joined: Fri Mar 13, 2009 7:15 am

Re: Interpreting efficiency measures for efficient designs

Post by johnr »

These are different efficiency measures I was playing around with in 2011 related to sampling error around the choice probabilities. I never did anything with these measures, and they remain available only in a research version of Ngene that hasn't been updated since 2013. I would suggest that you purchase a commercial version of the software as the libraries in the research version have substantially changed over the years, and the version you are using is not supported. Further, many of the features are related to our research at the time, most unpublished, and a lot of it still full of bugs.

John
tn73
Posts: 12
Joined: Thu May 16, 2013 9:29 pm

Re: Interpreting efficiency measures for efficient designs

Post by tn73 »

Thanks John, that explains everything!
Tom
Post Reply