forum.smartpls.com

Hi Samy, I'm glad that my comments made sense. However, the issue you raised regarding the bootstrap sample size kept bugging me. So I ran some tests that basically confirmed my hypothesis regarding the relationship between bootstrap sample size and stdev of bootstrap estimates (i.e. t-values): The ...

Hi Samy, well, interesting points. Here's what I think: 1) Bootstrap Sample Size I have not come across any recommendations regarding this parameter in the literature. However, your rationale for choosing an N smaller than the original sample doesn't convince me. Due to the random drawing of cases f...

From what I understand, those t-values reported directly in the PLS output are an experimental feature that can be safely ignored. I'm not aware of any useful application of these values.

Cheers,
Stefan

Hi Klaus, the number of cases in the original data set is (almost) irrelevant, since what you are trying to approximate with the bootstrap is the distribution of the parameter estimates. Significance is determined against an ordinary t-statistic table using the number of bootstrap runs as the df (i....

Dear PLS-Community, I would like to compare two subsamples of my data for differences in the structural path estimates. In particular, I would like to test, if the observed differences in the estimates are actually significant. Dibbern/ Chin (2005) describe a procedure using random permutations. How...

Klaus, you will be able to answer your questions on your own once you understand how the bootstrap procedure works: For each boostrap-run (100 runs in your case), the program selects a specified number of cases (75 in your case) at random from your sample with the possibility of drawing the same cas...

Christian, you are right that what I reported was simply "human error" (i.e. my stupidity) and not a bug. However, I would guess that I'm not the only one out there stupid enough to run into this (maybe others are simply not yet aware of it). What was confusing me (and maybe this is a sugg...

Thinking about it, I would actually prefer to see the t-values underneath each path coefficient rather than stars. Example:

(LV 1)
|
|
| 0.456
| (9.453)
|
v
(LV 2)

Cheers,
Stefan

Alex, Christian, I think I figured out what caused the differences in path estimates. It had to do with the treatment of missing values. Somehow, I had calclulated my models without the explicit missing value treatment (I thought the default was "replace with mean" as in SmartPLS 1.0) and ...

Ok, then it must have to do with the model specification in LVPLS. Let me run a couple more tests and if I still can't bring both models to the same estimates then I will send you all the files.

Cheers,
Stefan

All, after some more in-depth reading and some fooling around with Christian's Excel-sheets I was able to figure out a way to implement the procedure in a more flexible way. The sheets I created take MV- and LV-score output from PLS and calculate the cross-validated redundancy measure for a given va...

All, not sure really, if this is a bug or a feature... I've noticed that SmartPLS and LVPLS v1.8 give very different estimates for structural paths and indicator weights/loadings for models that are specified in an identical way and estimated using the same raw data. In some cases, the differences a...

I would argue for using the items directly as indicators in the PLS model.
However, I'm looking forward to comments from the more experienced researchers in this forum.
Cheers,
Stefan

Dirk,

basically, to test the predictive validity of structural PLS-models, you need to look at cross-validated redundancy (F-sq) of your main dependent variable(s).

F-sq < 0 indicates that your structural model has issues.

Cheers,
Stefan

I guess it is already on the todo-list for the next release, but supporting the modeling of moderation-/ interaction-effects would be a neat feature. While such modeling is already possible now, it is quite painstaking to generate the product-interaction indicators manually (especially, if you have ...