turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- where are my holdouts?

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

07-20-2010 06:23 PM

hey all,

i am currently trying to analyse my discrete choice data, and i was hoping to get some advise from you.

i created 16 holdout choice sets using kuhfeld's %mktex macro. They were tagged with the constant "w". Afterwards, transformed the linear design into a choice design, using %mktkey and %mktroll.

The problem I am facing now is the following:

Instead of my original72 choice sets, I have 88. However, i cannot identify which of the 88 choice sets is a holdout choice set, and which not, as data sets do not contain the indicating "w" constant anymore. I already collected my data.

I doubt that I could just randomly extract 16 choice sets and use them as holdouts. So - how can I find out which 16 of the 88 choice sets are the holdouts?

I really appreciate your help,

cheers from europe

i am currently trying to analyse my discrete choice data, and i was hoping to get some advise from you.

i created 16 holdout choice sets using kuhfeld's %mktex macro. They were tagged with the constant "w". Afterwards, transformed the linear design into a choice design, using %mktkey and %mktroll.

The problem I am facing now is the following:

Instead of my original72 choice sets, I have 88. However, i cannot identify which of the 88 choice sets is a holdout choice set, and which not, as data sets do not contain the indicating "w" constant anymore. I already collected my data.

I doubt that I could just randomly extract 16 choice sets and use them as holdouts. So - how can I find out which 16 of the 88 choice sets are the holdouts?

I really appreciate your help,

cheers from europe

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

Posted in reply to oraider

07-21-2010 04:59 AM

it might be of interest that the log provides the following warning after i transformed the linear into a choice design:

WARNING: The variable W is in the DESIGN= data set but not the KEY= data set.

maybe this helps.....

thanks

WARNING: The variable W is in the DESIGN= data set but not the KEY= data set.

maybe this helps.....

thanks