Bayesian variable selection in linear quantile mixed models for longitudinal data with application to macular degeneration.
Ontology highlight
ABSTRACT:
SUBMITTER: Ji Y
PROVIDER: S-EPMC7588124 | biostudies-literature | 2020
REPOSITORIES: biostudies-literature
Publications
Bayesian variable selection in linear quantile mixed models for longitudinal data with application to macular degeneration.
PloS one 20201026 10
This paper presents a Bayesian analysis of linear mixed models for quantile regression based on a Cholesky decomposition for the covariance matrix of random effects. We develop a Bayesian shrinkage approach to quantile mixed regression models using a Bayesian adaptive lasso and an extended Bayesian adaptive group lasso. We also consider variable selection procedures for both fixed and random effects in a linear quantile mixed model via the Bayesian adaptive lasso and extended Bayesian adaptive g ...[more]
Similar Datasets
Project description:This paper is concerned with the selection and estimation of fixed and random effects in linear mixed effects models. We propose a class of nonconcave penalized profile likelihood methods for selecting and estimating important fixed effects. To overcome the difficulty of unknown covariance matrix of random effects, we propose to use a proxy matrix in the penalized profile likelihood. We establish conditions on the choice of the proxy matrix and show that the proposed procedure enjoys the model selection consistency where the number of fixed effects is allowed to grow exponentially with the sample size. We further propose a group variable selection strategy to simultaneously select and estimate important random effects, where the unknown covariance matrix of random effects is replaced with a proxy matrix. We prove that, with the proxy matrix appropriately chosen, the proposed procedure can identify all true random effects with asymptotic probability one, where the dimension of random effects vector is allowed to increase exponentially with the sample size. Monte Carlo simulation studies are conducted to examine the finite-sample performance of the proposed procedures. We further illustrate the proposed procedures via a real data example.
| S-EPMC4026175 | biostudies-other
Project description:Quantile regression offers an invaluable tool to discern effects that would be missed by other conventional regression models, which are solely based on modeling conditional mean. Quantile regression for mixed-effects models has become practical for longitudinal data analysis due to the recent computational advances and the ready availability of efficient linear programming algorithms. Recently, quantile regression has also been extended to additive mixed-effects models, providing an efficient and flexible framework for nonparametric as well as parametric longitudinal forms of data analysis focused on features of the outcome beyond its central tendency. This study applies the additive quantile mixed model to analyze the longitudinal CD4 count of HIV-infected patients enrolled in a follow-up study at the Centre of the AIDS Programme of Research in South Africa. The objective of the study is to justify how the procedure developed can obtain robust nonlinear and linear effects at different conditional distribution locations. With respect to time and baseline BMI effect, the study shows a significant nonlinear effect on CD4 count across all fitted quantiles. Furthermore, across all fitted quantiles, the effect of the parametric covariates of baseline viral load, place of residence, and the number of sexual partners was found to be major significant factors on the progression of patients' CD4 count who had been initiated on the Highly Active Antiretroviral Therapy study.
| S-EPMC8429740 | biostudies-literature
Project description:A new generalized linear mixed quantile model for panel data is proposed. This proposed approach applies GEE with smoothed estimating functions, which leads to asymptotically equivalent estimation of the regression coefficients. Random effects are predicted by using the best linear unbiased predictors (BLUP) based on the Tweedie exponential dispersion distributions which cover a wide range of distributions, including those widely used ones, such as the normal distribution, Poisson distribution and gamma distribution. A Taylor expansion of the quantile estimating function is used to linearize the random effects in the quantile process. The parameter estimation is based on the Newton-Raphson iteration method. Our proposed quantile mixed model gives consistent estimates that have asymptotic normal distributions. Simulation studies are carried out to investigate the small sample performance of the proposed approach. As an illustration, the proposed method is applied to analyze the epilepsy data.
| S-EPMC7419003 | biostudies-literature
Project description:Age-related macular degeneration (AMD) is the leading cause of irreversible visual loss in the elderly in developed countries and typically affects more than 10% of individuals over age 80. AMD has a large genetic component, with heritability estimated to be between 45% and 70%. Numerous variants have been identified and implicate various molecular mechanisms and pathways for AMD pathogenesis but those variants only explain a portion of AMD's heritability. The goal of our study was to estimate the cumulative genetic contribution of common variants on AMD risk for multiple pathways related to the etiology of AMD, including angiogenesis, antioxidant activity, apoptotic signaling, complement activation, inflammatory response, response to nicotine, oxidative phosphorylation, and the tricarboxylic acid cycle. While these mechanisms have been associated with AMD in literature, the overall extent of the contribution to AMD risk for each is unknown.In a case-control dataset with 1,813 individuals genotyped for over 600,000 SNPs we used Genome-wide Complex Trait Analysis (GCTA) to estimate the proportion of AMD risk explained by SNPs in genes associated with each pathway. SNPs within a 50 kb region flanking each gene were also assessed, as well as more distant, putatively regulatory SNPs, based on DNaseI hypersensitivity data from ocular tissue in the ENCODE project.We found that 19 previously associated AMD risk SNPs contributed to 13.3% of the risk for AMD in our dataset, while the remaining genotyped SNPs contributed to 36.7% of AMD risk. Adjusting for the 19 risk SNPs, the complement activation and inflammatory response pathways still explained a statistically significant proportion of additional risk for AMD (9.8% and 17.9%, respectively), with other pathways showing no significant effects (0.3% - 4.4%).Our results show that SNPs associated with complement activation and inflammation significantly contribute to AMD risk, separately from the risk explained by the 19 known risk SNPs. We found that SNPs within 50 kb regions flanking genes explained additional risk beyond genic SNPs, suggesting a potential regulatory role, but that more distant SNPs explained less than 0.5% additional risk for each pathway.From these analyses we find that the impact of complement SNPs on risk for AMD extends beyond the established genome-wide significant SNPs.
| S-EPMC4606903 | biostudies-literature
Project description:Background Scleroderma is a serious chronic autoimmune disease in which a patient’s disease state manifests in several irregularly spaced longitudinal measures of lung, heart, skin, and other organ systems. Threshold crossings of pulmonary and cardiac measures indicate potentially life-threatening key clinical events including interstitial lung disease (ILD), cardiomyopathy, and pulmonary hypertension (PH). The statistical challenge is to accurately and precisely predict these events by using all of the clinical history for the patient at hand and for a reference population of patients. Methods We use a Bayesian mixed model approach to simultaneously characterize each individual’s future trajectories for several biomarkers. We estimate this model using a large population of patients from the Johns Hopkins Scleroderma Center Research Registry. The joint probabilities of critical lung and heart events are then calculated as a byproduct of the mixed model. Results The performance of this approach is substantially better than standard, more common alternatives. In order to predict an individual’s risks in a clinical setting, we also develop a cross-validated, sequential prediction (CVSP) algorithm. As additional data are observed during a patient’s visit, the algorithm sequentially produces updated predictions for the future longitudinal trajectories and for ILD, cardiomyopathy, and PH. The updated prediction distributions with little additional computing, for example within an electronic health record (EHR). Conclusions This method that generates real-time personalized risk estimates has been implemented within the electronic health record system for clinical testing. To our knowledge, this work represents the first approach to compute personalized risk estimates for multiple scleroderma complications. Supplementary Information The online version contains supplementary material available at 10.1186/s12874-021-01439-y.
| S-EPMC8590788 | biostudies-literature
Project description:Microorganisms play critical roles in human health and disease. They live in diverse communities in which they interact synergistically or antagonistically. Thus for estimating microbial associations with clinical covariates, such as treatment effects, joint (multivariate) statistical models are preferred. Multivariate models allow one to estimate and exploit complex interdependencies among multiple taxa, yielding more powerful tests of exposure or treatment effects than application of taxon-specific univariate analyses. Analysis of microbial count data also requires special attention because data commonly exhibit zero inflation, i.e., more zeros than expected from a standard count distribution. To meet these needs, we developed a Bayesian variable selection model for multivariate count data with excess zeros that incorporates information on the covariance structure of the outcomes (counts for multiple taxa), while estimating associations with the mean levels of these outcomes. Though there has been much work on zero-inflated models for longitudinal data, little attention has been given to high-dimensional multivariate zero-inflated data modeled via a general correlation structure. Through simulation, we compared performance of the proposed method to that of existing univariate approaches, for both the binary ("excess zero") and count parts of the model. When outcomes were correlated the proposed variable selection method maintained type I error while boosting the ability to identify true associations in the binary component of the model. For the count part of the model, in some scenarios the univariate method had higher power than the multivariate approach. This higher power was at a cost of a highly inflated false discovery rate not observed with the proposed multivariate method. We applied the approach to oral microbiome data from the Pediatric HIV/AIDS Cohort Oral Health Study and identified five (of 44) species associated with HIV infection.
| S-EPMC7308073 | biostudies-literature
Project description:We study the group bridge and the adaptive group bridge penalties for competing risks quantile regression with group variables. While the group bridge consistently identifies nonzero group variables, the adaptive group bridge consistently selects variables not only at group level but also at within-group level. We allow the number of covariates to diverge as the sample size increases. The oracle property for both methods is also studied. The performance of the group bridge and the adaptive group bridge is compared in simulation and in a real data analysis. The simulation study shows that the adaptive group bridge selects nonzero within-group variables more consistently than the group bridge. A bone marrow transplant study is provided as an example.
| S-EPMC5889760 | biostudies-literature
Project description:This paper develops a likelihood-based approach to analyze quantile regression (QR) models for continuous longitudinal data via the asymmetric Laplace distribution (ALD). Compared to the conventional mean regression approach, QR can characterize the entire conditional distribution of the outcome variable and is more robust to the presence of outliers and misspecification of the error distribution. Exploiting the nice hierarchical representation of the ALD, our classical approach follows a Stochastic Approximation of the EM (SAEM) algorithm in deriving exact maximum likelihood estimates of the fixed-effects and variance components. We evaluate the finite sample performance of the algorithm and the asymptotic properties of the ML estimates through empirical experiments and applications to two real life datasets. Our empirical results clearly indicate that the SAEM estimates outperforms the estimates obtained via the combination of Gaussian quadrature and non-smooth optimization routines of the Geraci and Bottai (2014) approach in terms of standard errors and mean square error. The proposed SAEM algorithm is implemented in the R package qrLMM().
| S-EPMC5667718 | biostudies-literature
Project description:Bayesian variable selection often assumes normality, but the effects of model misspecification are not sufficiently understood. There are sound reasons behind this assumption, particularly for large p: ease of interpretation, analytical and computational convenience. More flexible frameworks exist, including semi- or non-parametric models, often at the cost of some tractability. We propose a simple extension that allows for skewness and thicker-than-normal tails but preserves tractability. It leads to easy interpretation and a log-concave likelihood that facilitates optimization and integration. We characterize asymptotically parameter estimation and Bayes factor rates, under certain model misspecification. Under suitable conditions misspecified Bayes factors induce sparsity at the same rates than under the correct model. However, the rates to detect signal change by an exponential factor, often reducing sensitivity. These deficiencies can be ameliorated by inferring the error distribution, a simple strategy that can improve inference substantially. Our work focuses on the likelihood and can be combined with any likelihood penalty or prior, but here we focus on non-local priors to induce extra sparsity and ameliorate finite-sample effects caused by misspecification. We show the importance of considering the likelihood rather than solely the prior, for Bayesian variable selection. The methodology is in R package 'mombf'.
| S-EPMC6426142 | biostudies-literature
Project description:Generalized linear mixed models (GLMMs) continue to grow in popularity due to their ability to directly acknowledge multiple levels of dependency and model different data types. For small sample sizes especially, likelihood-based inference can be unreliable with variance components being particularly difficult to estimate. A Bayesian approach is appealing but has been hampered by the lack of a fast implementation, and the difficulty in specifying prior distributions with variance components again being particularly problematic. Here, we briefly review previous approaches to computation in Bayesian implementations of GLMMs and illustrate in detail, the use of integrated nested Laplace approximations in this context. We consider a number of examples, carefully specifying prior distributions on meaningful quantities in each case. The examples cover a wide range of data types including those requiring smoothing over time and a relatively complicated spline model for which we examine our prior specification in terms of the implied degrees of freedom. We conclude that Bayesian inference is now practically feasible for GLMMs and provides an attractive alternative to likelihood-based approaches such as penalized quasi-likelihood. As with likelihood-based approaches, great care is required in the analysis of clustered binary data since approximation strategies may be less accurate for such data.
| S-EPMC2883299 | biostudies-literature