17. Finite Mixture (FIMIX) Model in SmartPLS-4 || Dr. Dhaval Maheta

#sem, #smartpls, #construct, #latent, #model, #fimix, #observed, #unobserved, #heterogeneity, #segment, #smartpls4
Email: [e-mail]
Twitter: / dhavalmaheta77
LinkedIn: / dhaval-maheta-320200153
Facebook ID: / dhaval.maheta.37
Telegram ID: t.me/DhavalMaheta

[@michaelpaulse1]

Hello professor thank you for a great lesson. Just a question: The size of Partition 2 is less than the minimum sample determined by G Power, why is it then considered?

[@michaelpaulse1]

Could you send me your PPT as well? I did send you an email. Could you also send me your PPT on Micom? Kind regards, Michael PhD Student, South Africa

[@jazcpcuev9915]

Thanks is a ver very good lesson

[@ruchiragarai8227]

Hello sir, thankyou for this informative video. I have run fimix on a dataset, however after creation of the new data file, there is no Final Partition row present alongwith the other new FIMIX segment rows that were created. Does that mean the data does not have unobserved heterogeneity?

[@Franzi634]

Thank you! :) Just one question, why did you keep the mean replacement (Data) while running FIMIX? I thought it is recommanded to click on casewise deletion or am I wrong?

[@DhavalSaifaleeAaryash]

Casewise deletion our data quantity will reduce

[@cata.maican]

Hi,
in the last slide you mentioned something about quality standards for the partitions. Are these criteria the same as the step 2 (and 3) from MICOM/MGA? What happens if these standards are not met?
Thank you.

[@DhavalSaifaleeAaryash]

can you mention the time line

[@cata.maican]

Hi, about 38:40, step 4, the last part of the first bullet point, combined with the second bullet

[@DhavalSaifaleeAaryash]

@@cata.maican so this was related to reliability and validity of separate models you make considering fimix

[@lamthihoanganh406]

Thanks for your video; however, I have a question: How can I detect the second group for observed variables like Class Type? Unbalance percentage >> Observed variables?

[@DhavalSaifaleeAaryash]

you will have to check manually in spss or excel

[@lamthihoanganh406]

​@@DhavalSaifaleeAaryash Thanks for your prompt response but I don't know what criteria that we use to identify the Observed variable? Is it the same percentage?
For example, FIMIX-PLS group 1: 155 - 49.68%
FIMIX-PLS group 2: 157 - 50.32%
>> Gender:
Male: 229 - 73.4%
Female: 82 - 26.3%
>> Age Group
18 - 24: 57 - 18.3%
25 - 34: 179 - 57.4%
35 - 44: 69 - 22.1%
45 - 54: 7 - 2.2%
>> Marital Status
Single: 171 - 54.8%
Married: 138 - 44.2%
Others: 3 - 1%
>> We choose "Marital Status" right?

[@DhavalSaifaleeAaryash]

@@lamthihoanganh406 fimix is used for unobserved heterogeneity. So we use percentages

[@DhavalSaifaleeAaryash]

You can connect with me on mail: [e-mail]