1
|
Huo S, Morris JS, Zhu H. Ultra-Fast Approximate Inference Using Variational Functional Mixed Models. J Comput Graph Stat 2022; 32:353-365. [PMID: 37608921 PMCID: PMC10441618 DOI: 10.1080/10618600.2022.2107532] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/17/2021] [Accepted: 07/23/2022] [Indexed: 10/16/2022]
Abstract
While Bayesian functional mixed models have been shown effective to model functional data with various complex structures, their application to extremely high-dimensional data is limited due to computational challenges involved in posterior sampling. We introduce a new computational framework that enables ultra-fast approximate inference for high-dimensional data in functional form. This framework adopts parsimonious basis to represent functional observations, which facilitates efficient compression and parallel computing in basis space. Instead of performing expensive Markov chain Monte Carlo sampling, we approximate the posterior distribution using variational Bayes and adopt a fast iterative algorithm to estimate parameters of the approximate distribution. Our approach facilitates a fast multiple testing procedure in basis space, which can be used to identify significant local regions that reflect differences across groups of samples. We perform two simulation studies to assess the performance of approximate inference, and demonstrate applications of the proposed approach by using a proteomic mass spectrometry dataset and a brain imaging dataset. Supplementary materials are available online.
Collapse
Affiliation(s)
| | - Jeffrey S Morris
- Department of Biostatistics, Epidemiology and Informatics, Department of Statistics, University of Pennsylvania
| | | |
Collapse
|
2
|
Degani E, Maestrini L, Toczydłowska D, Wand MP. Sparse linear mixed model selection via streamlined variational Bayes. Electron J Stat 2022. [DOI: 10.1214/22-ejs2063] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
Affiliation(s)
- Emanuele Degani
- Dipartimento di Scienze Statistiche Università degli Studi di Padova, Padova, Italy
| | - Luca Maestrini
- Research School of Finance, Actuarial Studies and Statistics The Australian National University, Canberra, Australia
| | - Dorota Toczydłowska
- School of Mathematical and Physical Sciences University of Technology Sydney, Sydney, Australia
| | - Matt P. Wand
- School of Mathematical and Physical Sciences University of Technology Sydney, Sydney, Australia
| |
Collapse
|
3
|
Laplace approximations for fast Bayesian inference in generalized additive models based on P-splines. Comput Stat Data Anal 2021. [DOI: 10.1016/j.csda.2020.107088] [Citation(s) in RCA: 6] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
|
4
|
Lee JYL, Green PJ, Ryan LM. Analysis of grouped data using conjugate generalized linear mixed models. Biometrika 2019. [DOI: 10.1093/biomet/asz053] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022] Open
Abstract
Summary
This article concerns a class of generalized linear mixed models for two-level grouped data, where the random effects are uniquely indexed by groups and are independent. We derive necessary and sufficient conditions for the marginal likelihood to be expressed in explicit form. These models are unified under the conjugate generalized linear mixed models framework, where conjugate refers to the fact that the marginal likelihood can be expressed in closed form, rather than implying inference via the Bayesian paradigm. The proposed framework allows simultaneous conjugacy for Gaussian, Poisson and gamma responses, and thus can accommodate both unit- and group-level covariates. Only group-level covariates can be incorporated for the binomial distribution. In a simulation of Poisson data, our framework outperformed its competitors in terms of computational time, and was competitive in terms of robustness against misspecification of the random effects distributions.
Collapse
Affiliation(s)
- Jarod Y L Lee
- School of Mathematical and Physical Sciences, University of Technology Sydney, Sydney, Australia
| | - Peter J Green
- School of Mathematical and Physical Sciences, University of Technology Sydney, Sydney, Australia
| | - Louise M Ryan
- School of Mathematical and Physical Sciences, University of Technology Sydney, Sydney, Australia
| |
Collapse
|
5
|
Talagala PD, Hyndman RJ, Smith-Miles K, Kandanaarachchi S, Muñoz MA. Anomaly Detection in Streaming Nonstationary Temporal Data. J Comput Graph Stat 2019. [DOI: 10.1080/10618600.2019.1617160] [Citation(s) in RCA: 22] [Impact Index Per Article: 4.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/26/2022]
Affiliation(s)
- Priyanga Dilini Talagala
- Department of Econometrics and Business Statistics, Monash University, Clayton, Victoria, Australia
- ARC Centre of Excellence for Mathematics and Statistical Frontiers (ACEMS), Australia
| | - Rob J. Hyndman
- Department of Econometrics and Business Statistics, Monash University, Clayton, Victoria, Australia
- ARC Centre of Excellence for Mathematics and Statistical Frontiers (ACEMS), Australia
| | - Kate Smith-Miles
- ARC Centre of Excellence for Mathematics and Statistical Frontiers (ACEMS), Australia
- School of Mathematics and Statistics, University of Melbourne, Parkville, Victoria, Australia
| | - Sevvandi Kandanaarachchi
- Department of Econometrics and Business Statistics, Monash University, Clayton, Victoria, Australia
- ARC Centre of Excellence for Mathematics and Statistical Frontiers (ACEMS), Australia
| | - Mario A. Muñoz
- ARC Centre of Excellence for Mathematics and Statistical Frontiers (ACEMS), Australia
- School of Mathematics and Statistics, University of Melbourne, Parkville, Victoria, Australia
| |
Collapse
|
6
|
Hui FKC, You C, Shang HL, Müller S. Semiparametric Regression Using Variational Approximations. J Am Stat Assoc 2019. [DOI: 10.1080/01621459.2018.1518235] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/28/2022]
Affiliation(s)
- Francis K. C. Hui
- Mathematical Sciences Institute, The Australian National University, Canberra, Australia
| | - C. You
- School of Mathematical Sciences, University of Nottingham, Ningbo, China
| | - H. L. Shang
- Research School of Finance, Actuarial Studies, and Statistics, The Australian National University, Canberra, Australia
| | - Samuel Müller
- School of Mathematics and Statistics, University of Sydney, Sydney, Australia
| |
Collapse
|
7
|
Kim AS, Wand MP. On expectation propagation for generalised, linear and mixed models. AUST NZ J STAT 2018. [DOI: 10.1111/anzs.12199] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
Affiliation(s)
- Andy S.I. Kim
- University of Technology Sydney and Australian Research Council Centre of Excellence for Mathematical and Statistical Frontiers
- School of Mathematical and Physical Sciences; University of Technology Sydney; Broadway 2007 Australia
| | - Matt P. Wand
- University of Technology Sydney and Australian Research Council Centre of Excellence for Mathematical and Statistical Frontiers
- School of Mathematical and Physical Sciences; University of Technology Sydney; Broadway 2007 Australia
| |
Collapse
|
8
|
Wand MP. Fast Approximate Inference for Arbitrarily Large Semiparametric Regression Models via Message Passing. J Am Stat Assoc 2017. [DOI: 10.1080/01621459.2016.1197833] [Citation(s) in RCA: 14] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/21/2022]
Affiliation(s)
- M. P. Wand
- School of Mathematical and Physical Sciences, University of Technology Sydney, Sydney, Australia, and Australian Research Council Centre of Excellence for Mathematical and Statistical Frontiers, Queensland University of Technology (QUT), Brisbane, Australia
| |
Collapse
|
9
|
Ong VMH, Mensah DK, Nott DJ, Jo S, Park B, Choi T. A variational Bayes approach to a semiparametric regression using Gaussian process priors. Electron J Stat 2017. [DOI: 10.1214/17-ejs1324] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
|
10
|
Lee CYY, Wand MP. Streamlined mean field variational Bayes for longitudinal and multilevel data analysis. Biom J 2016; 58:868-95. [DOI: 10.1002/bimj.201500007] [Citation(s) in RCA: 16] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/11/2015] [Revised: 12/06/2015] [Accepted: 12/18/2015] [Indexed: 11/09/2022]
Affiliation(s)
- Cathy Yuen Yi Lee
- School of Mathematical and Physical Sciences University of Technology Sydney P.O. Box 123 Broadway New South Wales 2007 Australia
| | - Matt P. Wand
- School of Mathematical and Physical Sciences University of Technology Sydney P.O. Box 123 Broadway New South Wales 2007 Australia
| |
Collapse
|
11
|
Tran MN, Nott DJ, Kuk AYC, Kohn R. Parallel Variational Bayes for Large Datasets With an Application to Generalized Linear Mixed Models. J Comput Graph Stat 2016. [DOI: 10.1080/10618600.2015.1012293] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/23/2022]
|
12
|
Luedtke AR, van der Laan MJ. STATISTICAL INFERENCE FOR THE MEAN OUTCOME UNDER A POSSIBLY NON-UNIQUE OPTIMAL TREATMENT STRATEGY. Ann Stat 2016; 44:713-742. [PMID: 30662101 PMCID: PMC6338452 DOI: 10.1214/15-aos1384] [Citation(s) in RCA: 66] [Impact Index Per Article: 8.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
Abstract
We consider challenges that arise in the estimation of the mean outcome under an optimal individualized treatment strategy defined as the treatment rule that maximizes the population mean outcome, where the candidate treatment rules are restricted to depend on baseline covariates. We prove a necessary and sufficient condition for the pathwise differentiability of the optimal value, a key condition needed to develop a regular and asymptotically linear (RAL) estimator of the optimal value. The stated condition is slightly more general than the previous condition implied in the literature. We then describe an approach to obtain root-n rate confidence intervals for the optimal value even when the parameter is not pathwise differentiable. We provide conditions under which our estimator is RAL and asymptotically efficient when the mean outcome is pathwise differentiable. We also outline an extension of our approach to a multiple time point problem. All of our results are supported by simulations.
Collapse
|
13
|
Lee CYY, Wand MP. Variational methods for fitting complex Bayesian mixed effects models to health data. Stat Med 2016; 35:165-88. [PMID: 26415742 DOI: 10.1002/sim.6737] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/07/2015] [Revised: 08/05/2015] [Accepted: 08/27/2015] [Indexed: 11/10/2022]
Abstract
We consider approximate inference methods for Bayesian inference to longitudinal and multilevel data within the context of health science studies. The complexity of these grouped data often necessitates the use of sophisticated statistical models. However, the large size of these data can pose significant challenges for model fitting in terms of computational speed and memory storage. Our methodology is motivated by a study that examines trends in cesarean section rates in the largest state of Australia, New South Wales, between 1994 and 2010. We propose a group-specific curve model that encapsulates the complex nonlinear features of the overall and hospital-specific trends in cesarean section rates while taking into account hospital variability over time. We use penalized spline-based smooth functions that represent trends and implement a fully mean field variational Bayes approach to model fitting. Our mean field variational Bayes algorithms allow a fast (up to the order of thousands) and streamlined analytical approximate inference for complex mixed effects models, with minor degradation in accuracy compared with the standard Markov chain Monte Carlo methods.
Collapse
Affiliation(s)
- Cathy Yuen Yi Lee
- School of Mathematical and Physical Sciences, University of Technology Sydney, Ultimo, New South Wales, 2007, Australia
| | - Matt P Wand
- School of Mathematical and Physical Sciences, University of Technology Sydney, Ultimo, New South Wales, 2007, Australia
| |
Collapse
|
14
|
Kim ASI, Wand MP. The explicit form of expectation propagation for a simple statistical model. Electron J Stat 2016. [DOI: 10.1214/16-ejs1114] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
|
15
|
Menictas M, Wand MP. Variational Inference for Heteroscedastic Semiparametric Regression. AUST NZ J STAT 2015. [DOI: 10.1111/anzs.12105] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
Affiliation(s)
- Marianne Menictas
- School of Mathematical Sciences; University of Technology Sydney; Broadway 2007 Australia
| | - Matt P. Wand
- School of Mathematical Sciences; University of Technology Sydney; Broadway 2007 Australia
| |
Collapse
|
16
|
Variational inferences for partially linear additive models with variable selection. Comput Stat Data Anal 2014. [DOI: 10.1016/j.csda.2014.07.003] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022]
|