Search found 7 matches
- Thu Jun 04, 2009 7:42 pm
- Forum: SmartPLS 3 - Requested Features
- Topic: Enhanced support of time series
- Replies: 0
- Views: 3662
Enhanced support of time series
Hello SmartPLS team, I'm working in Econometrics with time series data, that means the observations are ordered by e.g. months or years. An important aspect is the use of lagged explaining variables. If I want to use lagged variables I have to manipulate the data set by hand with an external editor ...
- Fri May 29, 2009 12:51 pm
- Forum: Method and application
- Topic: bootstraping t-values: t oder normal distribution
- Replies: 1
- Views: 2341
bootstraping t-values: t oder normal distribution
Hello,
is it recommend to assume the student t-distribution or the normal distibution for the bootstrapping t-values.
What is the appropriate choice for the degrees of freedom, the sample size or the number of bootstraps?
Thanks!
is it recommend to assume the student t-distribution or the normal distibution for the bootstrapping t-values.
What is the appropriate choice for the degrees of freedom, the sample size or the number of bootstraps?
Thanks!
- Sun Jul 29, 2007 5:51 pm
- Forum: Method and application
- Topic: Positive Path Coefficient vs negative Correlation
- Replies: 13
- Views: 19587
This is still a little bit confusing. To make it clear: a negative path coefficient with a positive correlation still indicates a neg. relationship of the 2 LV? This bothers me because in multiple linear regression coefficients and correlation are identical (as long as you use standarized data) or a...
- Sat Jul 28, 2007 1:42 pm
- Forum: SmartPLS 2 - FAQ
- Topic: reflective or formative LV as default?
- Replies: 2
- Views: 2748
- Sat Jul 28, 2007 1:41 pm
- Forum: Method and application
- Topic: Positive Path Coefficient vs negative Correlation
- Replies: 13
- Views: 19587
Positive Path Coefficient vs negative Correlation
Hi everyone, I have a PLS modell with 8 LV and several pathes between them. At some pathes I receive a positive Path Coefficient (which is significant), but a negative Correlation ot the LV and vice versa. 1. So how can the Path Coefficients be interpreted, similar to the Coefficients in the simple ...
- Fri Jul 27, 2007 9:38 pm
- Forum: SmartPLS 2 - FAQ
- Topic: reflective or formative LV as default?
- Replies: 2
- Views: 2748
reflective or formative LV as default?
Hi
As default the arrows point from the latent variable to the manifest variables, is that supposed to mean reflective? Changing to formative can be done with the "invert measurement model", is that right?
Thanks
As default the arrows point from the latent variable to the manifest variables, is that supposed to mean reflective? Changing to formative can be done with the "invert measurement model", is that right?
Thanks
- Thu Mar 15, 2007 2:58 am
- Forum: SmartPLS 2 - FAQ
- Topic: Extract more than 1 latent Variable
- Replies: 1
- Views: 2191
Extract more than 1 latent Variable
Hi,
I just found that work on PLS with a small example:
http://www.utdallas.edu/~herve/Abdi-PLSR2007-pretty.pdf
The Authors extracts more than one LV from one Block of manifest variables. Is that possible in Smart-PLS, too?
thanks a lot
I just found that work on PLS with a small example:
http://www.utdallas.edu/~herve/Abdi-PLSR2007-pretty.pdf
The Authors extracts more than one LV from one Block of manifest variables. Is that possible in Smart-PLS, too?
thanks a lot