Conjoint Analysis-Holdout cases

291 views
Skip to first unread message

Makarand Wath

unread,
Feb 23, 2023, 12:19:36 AM2/23/23
to dataanalys...@googlegroups.com
Dear Members,

Are there any criteria for adding the number of holdout cases in conjoint analysis?
(want to know the number of holdout cases required in conjoint analysis?)

Makarand Wath (Dr.)

Neeraj Kaushik

unread,
Feb 23, 2023, 8:19:36 PM2/23/23
to dataanalys...@googlegroups.com
Holdout cases, also known as validation cases or validation respondents, are a subset of respondents in a conjoint analysis who are not used in the estimation process but are instead held back for the purpose of validating the accuracy of the model. The purpose of using holdout cases is to ensure that the model developed from the analysis is able to accurately predict the preferences of new customers.

The number of holdout cases required in a conjoint analysis depends on various factors such as the sample size, the complexity of the analysis, and the research objectives. As a general rule of thumb, the number of holdout cases should be at least 10% of the sample size, but it can be increased if the research requires more rigorous validation.

Some criteria that can be used to determine the number of holdout cases required in a conjoint analysis include:

Statistical significance: The number of holdout cases should be large enough to provide statistically significant results. The sample size and the complexity of the analysis can affect the statistical significance of the results.

Research objectives: The number of holdout cases should be based on the research objectives. If the research requires a high level of accuracy, then more holdout cases may be needed.

Data quality: The number of holdout cases should be based on the quality of the data. If the data is of high quality, then fewer holdout cases may be needed.

Model complexity: The number of holdout cases should be based on the complexity of the model. If the model is complex, then more holdout cases may be needed.

In summary, the number of holdout cases required in a conjoint analysis depends on several factors and should be determined based on the research objectives, data quality, statistical significance, and model complexity. A general guideline is to have at least 10% of the sample size as holdout cases.

Source: CHATGPT

--
Protocols of this Group:
 
1. Plz search previous post in group before posing the question.
2. Don't write query in someone's post. Always use the option of New topic for the new question. You can do this by writing to dataanalys...@googlegroups.com
3. Its better to give a proper subject to your post/query. It'll help others while searching.
4. Never write Open ended queries. This group intend to help research scholars NOT FOR WORK THEM.
5. Never write words like URGENT in ur posts. People will help them when they are free.
6. Never upload any info about National Seminars/Conferences. Send such info on personal emails. And feel free to share any RESEARCH related info.
7. No Happy New Year, Happy Diwali, Happy Holi, Happy B'day, Happy Anniversary etc. allowed on this group.
8. Few months back there was a facility for asking & sharing the Research Papers. Now there is no provision of asking for the research paper here.
 
Let’s make a better research environment.
---
You received this message because you are subscribed to the Google Groups "DataAnalysis" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dataanalysistrai...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dataanalysistraining/CAFiJzvCkW%2BZahV98-Oc5jX1dpvr%2BBFuyq6uPdREwvGuhfiLqRQ%40mail.gmail.com.

Mi Nam

unread,
Aug 25, 2023, 7:27:27 PM8/25/23
to DataAnalysis

Dear Members,
I am running conjoint analysis with the following command:

ONJOINT PLAN = '/Users/mac/congtyx1.sav'
                     /DATA = '/Users/mac/khaosat.sav'
                     /SEQUENCE = Sanpham1 TO Sanpham11
                     /SUBJECT = Respondents
                     /FACTORS= Huongvi (DISCRETE) Thuonghieu (DISCRETE) Chatluong (LINEAR MORE) Xuatxu (DISCRETE) Loairuouvang (DISCRETE)
   /PRINT=SUMMARYONLY.

then it said "Factors are not all orthogonal" what do I have to fix? thanks everyone
Vào lúc 08:19:36 UTC+7 ngày Thứ Sáu, 24 tháng 2, 2023, Neeraj Kaushik đã viết:
Screenshot 2023-08-26 at 12.48.32 AM.png
Reply all
Reply to author
Forward
0 new messages