Hi smartPLS-team,
what is the reasoning behind the pre-segmentation option? Is it to reduce calculation time but at the same time increasing the risk to getting stuck in a local optimum?
Thanks for your support!
Jan
PLS-POS: Pre-segmentation setting
-
- PLS Expert User
- Posts: 116
- Joined: Fri Sep 12, 2014 2:12 pm
- Real name and title: Jan Schreier
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: PLS-POS: Pre-segmentation setting
Yes, more or less this is a good summary.
With the pre-segmentation, in the first round all observations are switched to the group where they have the minimum distance. Later in the algorithm, we only switch one observation at a time. The reason is that switching to many observations at a time (like in PLS-REBUS) leads to very unstable groups and therefore imperfect assignments. In the first round, however, it might give the algorithm a boost in terms of avoiding some otherwise necessary iterations.
With the pre-segmentation, in the first round all observations are switched to the group where they have the minimum distance. Later in the algorithm, we only switch one observation at a time. The reason is that switching to many observations at a time (like in PLS-REBUS) leads to very unstable groups and therefore imperfect assignments. In the first round, however, it might give the algorithm a boost in terms of avoiding some otherwise necessary iterations.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
-
- PLS Expert User
- Posts: 116
- Joined: Fri Sep 12, 2014 2:12 pm
- Real name and title: Jan Schreier
Re: PLS-POS: Pre-segmentation setting
Thank you for the answer, Jan-Michael!
Do you have or know of any empirical tests of this pre-segmentation that give some rules of thumb if one should use it or not?
Do you have or know of any empirical tests of this pre-segmentation that give some rules of thumb if one should use it or not?
-
- SmartPLS Developer
- Posts: 1284
- Joined: Tue Mar 28, 2006 11:09 am
- Real name and title: Dr. Jan-Michael Becker
Re: PLS-POS: Pre-segmentation setting
From our simulation work when we evaluated the method we usually used it because it didn't reduced quality of the results, but improved performance.
Dr. Jan-Michael Becker, BI Norwegian Business School, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de
-
- PLS Expert User
- Posts: 116
- Joined: Fri Sep 12, 2014 2:12 pm
- Real name and title: Jan Schreier
Re: PLS-POS: Pre-segmentation setting
Thanks again!