Some questions about Partial Least Squares SEM in R

General
1

Hi everyone,

I am using the plspm package to conduct a Partial Least Squares SEM in R for my path model. I have three questions I would like to seek your advice as follows.

  1. How can I specify that I want to run 1000 bootstrap samples? It appears that the default command "boot.val = TRUE" will only run 500 samples.

  2. Is there any command that can help me compare the strength of two effects in my path model? For example, let's say I want to know whether IV X1 has a stronger association with DV Y than IV X2, given that X1 and X2 have the same measurement scale.

  3. If #2 is not possible, I plan to save the data of all bootstrap samples along with the path coefficients, so that I can manually construct the difference between effects and its bootstrapped confidence intervals to run the comparison. For this approach, is there any command that can allow me to save the data of the bootstrapped samples?

Thank you very much for your help!

2

according to the documentation, the default is 100; and to change this you would add in the br param,
i.e.

plsms( ...  , # your other things here
             boot.val = TRUE,br=1000)
3

Thank you very much! May I ask for your input regarding the two remaining matters as well?

4

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.