University of Twente Student Theses
The Bayesian Covariance Structure Growth Model for Intensive and Longitudinal Data
Hoopen, T. ten (2024) The Bayesian Covariance Structure Growth Model for Intensive and Longitudinal Data.
|
PDF
145kB |
| Abstract: | The linear mixed-effect (LME) model, as a growth model, is compared to the Bayesian covariance structure model (BCSM). The BCSM has numerous advantages over standard growth and linear mixed-effect models. One advantage is that the BCSM allows the random factor variance to get close to zero. The performance of the BCSM for growth modeling applications on longitudinal big data is examined in light of decreased random factor variance. A simulation study was conducted to examine the performance of the BCSM against the LME model. Both regular-sized data and big data were examined. Results indicate that when the random factor variance was minimal, the BCSM outperformed the LME model. However, a higher number of data iterations is needed to examine the performance of the BCSM against even lower levels of random factor variance. Furthermore, the BCSM should be tested on data consisting of higher-order polynomial terms as is common in growth models. In conclusion, the report indicates that the BCSM outperforms the LME model in addressing lower levels of random factor variance. Additionally, it can be stated that the BCSM is an effective model for estimating both fixed effects and reducing random factor variance when handling intensive longitudinal data. |
| Item Type: | Essay (Bachelor) |
| Faculty: | BMS: Behavioural, Management and Social Sciences |
| Subject: | 70 social sciences in general, 77 psychology |
| Programme: | Psychology BSc (56604) |
| Link to this item: | https://purl.utwente.nl/essays/100118 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page