Português
Bayesian structural equation modeling (BSEM) is receiving increasing interest mainly due to its capability of solving some of the issues found in the mainstream frequentist approach (e.g., nonconvergence, Heywood cases, sample size, inadmissible solutions) and because it allows fitting complex models that classical maximum likelihood methods might struggle to fit (Merkle & Rosseel, 2018). However, BSEM can be computationally intensive. In fact, the computational cost of the Bayesian analysis harmed the more frequent use of Bayesian statistics. The use of Bayesian methods has improved, mainly due to the computational advances, presenting researchers with more flexible, and powerful tools. Nowadays, Bayesian analysis is an established branch of methodology for model estimation (van de Schoot et al., 2021). In part, this is due to two aspects: the increased popularity of Bayesian methodology, and the advent of Markov chain Monte Carlo (MCMC) methods (Depaoli, 2021). Bayesian statistics benefit from MCMC methods since Bayesian analysis heavily relies on multidimensional integration. MCMC comprises a set of computational algorithms that can help to solve high‐dimensional, and complex modeling situations (South et al., 2022). MCMC can help Bayesian statistics by — for example — reconstructing the posterior distribution (Depaoli, 2021). MCMC can benefit greatly from a parallel computing environment, which allows to perform extensive calculations simultaneously. The advances in consumer computer hardware make parallel computing widely available to most users. Many computer video graphic cards support parallel computing. The use of the graphics processing units (GPU) usually provide meaningful gains in terms of performance (Češnovar et al., 2019). There is no doubt that parallel computing is of deep importance, the use of this technology can greatly improve several fields of statistics. One of those fields is BSEM, as so, the objective of this project is to understand the extent to which GPU processing gains make it worth in BSEM.
| Research Centre | Research Group | Role in Project | Begin Date | End Date |
|---|---|---|---|---|
| BRU-Iscte | Data Analytics | Leader | 2022-03-30 | 2022-09-30 |
| Institution | Country | Role in Project | Begin Date | End Date |
|---|---|---|---|---|
| University of Missouri (MU) | United States of America | Partner | 2022-03-30 | 2022-09-30 |
| Name | Affiliation | Role in Project | Begin Date | End Date |
|---|---|---|---|---|
| Jorge Sinval | Integrated Researcher (BRU-Iscte); | Principal Researcher | 2022-03-30 | 2022-09-30 |
| Reference/Code | Funding DOI | Funding Type | Funding Program | Funding Amount (Global) | Funding Amount (Local) | Begin Date | End Date |
|---|---|---|---|---|---|---|---|
| CPCA/A1/435377/2021 | -- | Award | FCT - FCT/CPCA/2021/01 - Portugal | 823 | 823 | 2022-03-30 | 2022-09-30 |
No records found.
No records found.
No records found.
No records found.
With the objective to increase the research activity directed towards the achievement of the United Nations 2030 Sustainable Development Goals, the possibility of associating scientific projects with the Sustainable Development Goals is now available in Ciência_Iscte. These are the Sustainable Development Goals identified for this project. For more detailed information on the Sustainable Development Goals, click here.
