Page 1 of 1

Constraint PLS-POS / FIMIX Results to values >=0

Posted: Thu Sep 03, 2015 7:02 am
by janschreier
Hahn et al. 2002 write "Our initial disaggregate solutions contained some small, negative coefficients for the drivers. With the assumption of independence, an interpretation would be difficult. For example, a negative impact of service to satisfaction would mean that higher quality service lowers overall satisfaction. To prevent such non-interpretable solutions, we constrained our coefficients to be equal or higher than zero. Consequently we only obtain local but interpretable optima for our coefficients."

Would love to see such an option in smartPLS (also for PLS-POS if possible) or are the reasons why this might be a bad idea?!

Thanks in advance!

Jan

--------
Hahn, C., Johnson, M.D., Herrmann, A., Huber, F., 2002. Capturing customer heterogeneity using a finite mixture PLS approach. Schmalenbach Business Review 54, 243–269.

Re: Constraint PLS-POS / FIMIX Results to values >=0

Posted: Thu Sep 03, 2015 6:03 pm
by cringle
Suggestions how to implement constraints in the PLS-SEM algorithm and PLS segmentation methods are welcome :))

This is definately a hot topic which some research groups try to solve.

At the moment, when you let the data speak this can always happen and is nothing unusual (at least in my many segmentation studies).

Best
CR

Re: Constraint PLS-POS / FIMIX Results to values >=0

Posted: Thu Sep 03, 2015 6:13 pm
by janschreier
i will have a deeper look into your publication list then (not that I haven't read quite a bit yet...)
for my PhD I am still trying to keep my hands clean of too crazy math stuff - so if I post a suggestion here you know that I failed in the prior matter ;)

Re: Constraint PLS-POS / FIMIX Results to values >=0

Posted: Thu Sep 03, 2015 9:11 pm
by janschreier
I just went through your recent papers and did not find anything on negative path coefficients. Did you refer to your professional studies? If this is also something you dealt with and reported in a published study I'd appreciate the study's title,

Merci beaucoup!
Jan

Re: Constraint PLS-POS / FIMIX Results to values >=0

Posted: Fri Sep 04, 2015 5:43 am
by cringle
Here they are:

Sarstedt Marko, Schwaiger Manfred, Ringle Christian M. Do We Fully Understand the Critical Success Factors of Customer Satisfaction with Industrial Goods? - Extending Festge and Schwaiger’s Model to Account for Unobserved Heterogeneity. Journal of Business Market Management 2009; 3 (3): 185-206.

Ringle Christian M., Sarstedt Marko, Mooi Erik A. Response-Based Segmentation Using Finite Mixture Partial Least Squares: Theoretical Foundations and an Application to American Customer Satisfaction Index Data. Annals of Information Systems 2010; 8: 19-49.

and to some extend this one

Sarstedt Marko, Ringle Christian M. Treating Unobserved Heterogeneity in PLS Path Modelling: A Comparison of FIMIX-PLS with Different Data Analysis Strategies. Journal of Applied Statistics 2010; 37 (8): 1299-1318.

Re: Constraint PLS-POS / FIMIX Results to values >=0

Posted: Fri Sep 04, 2015 10:49 pm
by janschreier
Thank you!