Hi!
I hope someone can help me out and tell me how to deal with two independent variables (LV's) which I believe are correlated. Is there any way to take account of this correlation without drawing a path and thus implying a cause-effect relationship? I tried drawing a two paths (implying a recursive relationship...) but PLS does not accept that...
Thanks!
recursive relationships and correlations
-
- PLS Junior User
- Posts: 1
- Joined: Wed Nov 02, 2005 4:50 pm
- Real name and title:
- Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Re: recursive relationships and correlations
In LISREL (covariance based) you must draw the covariance, but in PLS you don´t need.
See viewtopic.php?t=125
IN the output you could see de matrix of correlations between LVs.
PLS minimize the error´s covariance (predictive efficient), LISREL minimize the diference between data and predicted matrix of covariance.
Best regards.
See viewtopic.php?t=125
IN the output you could see de matrix of correlations between LVs.
PLS minimize the error´s covariance (predictive efficient), LISREL minimize the diference between data and predicted matrix of covariance.
Best regards.
Prof. Dr. Diogenes de Souza Bido