Page 1 of 1

Time series data

Posted: Fri Nov 11, 2016 1:13 pm
by laurathiel
Hello everyone,

I have a small problem and I hope that someone can help me. I have time series data which I tried to analyse with SmartPls. The problem is that I don't know who to control the factor "trend". I have choosen data for health care, education and economy from 2000 - 2015. The values of health care are decreasing in this time frame the values of economy are increasing. Can I add a latent variable for time? Or is there any other solution how I can control this?

I would be thankful for any idea or help.

Re: Time series data

Posted: Thu Mar 02, 2017 11:09 am
by cringle
PLS and time series data is a underresearched topic. So far, I have only seen SmartPLS applications that estimate the PLS path model with different datasets from different points of time.

Please share if you further advance this issues.

Best regards
Christian

Re: Time series data

Posted: Sat Feb 22, 2020 12:15 pm
by Ingrid Snijders
Dear Professor Ringle,

I want to apply the PLS-SEM approach for a CLPA study with data from two time points, where I investigate the contribution of Relationship Quality (five dimensions that form a HOC) on Student Engagement and Student Loyalty. In two other studies, one among enrolled higher education students , and one among alumni, I've already successfully applied PLS-SEM using SmartPLS. However, this was cross-sectional data.

Going through the literature, mostly, CB-SEM using MPlus is suggested. Within the forum, I saw an older post on time series and I can't find an exact answer whether PLS-SEM (and if so in what way) is suitable for such a study. I am still in favor of applying the PLS-SEM approach because I know this method (and I think I've already found significant findings). Furthermore, the research's aim is exploration, and for the CLPA study to examine temporality of RQ (dimensions), and, directionality/cause-and-effect relations.

What do you recommend?

If you need more information please let me know. Looking forward to your reply.

Best,

Ingrid Snijders