Please use this identifier to cite or link to this item:
http://hdl.handle.net/10077/30908
Title: | Streamlined Variational Inference for Linear Mixed Models with Crossed Random Effects | Authors: | Menictas, Marianne Di Credico, Gioia Wand, Matt P. |
Keywords: | Mean field variational Bayes; Item response theory; Rasch analysis; Scalable statistical methodology; Sparse least squares systems | Issue Date: | 2020 | Publisher: | EUT Edizioni Università di Trieste | Source: | Marianne Menictas, Gioia Di Credico, Matt P. Wand, "Streamlined Variational Inference for Linear Mixed Models with Crossed Random Effects", Trieste, EUT Edizioni Università di Trieste, 2020 | Series/Report no.: | DEAMS Research Paper Series, 2020, 2 | Pages: | 34 | Abstract: | We derive streamlined mean field variational Bayes algorithms for fitting linear mixed models with crossed random effects. In the most general situation, where the dimensions of the crossed groups are arbitrarily large, streamlining is hindered by lack of sparseness in the underlying least squares system. Because of this fact we also consider a hierarchy of relaxations of the mean field product restriction. The least stringent product restriction delivers a high degree of inferential accuracy. However, this accuracy must be mitigated against its higher storage and computing demands. Faster sparse storage and computing alternatives are also provided, but come with the price of diminished inferential accu-racy. This article provides full algorithmic details of three variational inference strategies, presents detailed empirical results on their pros and cons and, thus, guides the users on their choice of variational inference approach depending on the problem size and computing resources. |
Type: | Book | URI: | http://hdl.handle.net/10077/30908 | eISBN: | 978-88-5511-190-4 | Rights: | Attribution-NonCommercial-NoDerivatives 4.0 Internazionale |
Appears in Collections: | DEAMS Research Paper Series 2020, 2 |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
DEAMS_RP_2020_2_Menictas_DiCredico_Wand.pdf | Full Text | 1.1 MB | Adobe PDF | ![]() View/Open |
DEAMS_RP_2020_2_Front_cover.jpg | Front Cover | 213.03 kB | JPEG | ![]() View/Open |
CORE Recommender
Page view(s)
145
checked on Mar 29, 2023
Download(s)
121
checked on Mar 29, 2023
Google ScholarTM
Check
This item is licensed under a Creative Commons License