-
Prospective Prediction of Body Mass Index Trajectories using Multi-task Gaussian Processes
Authors:
Arthur Leroy,
Varsha Gupta,
Mya Thway Tint,
Delicia Ooi Shu Qin,
Keith M. Godfrey,
Fabian Yap,
Leck Ngee,
Yung Seng Lee,
Johan G. Eriksson,
Navin Michael,
Mauricio A. Alvarez,
Dennis Wang
Abstract:
Clinicians often investigate the body mass index (BMI) trajectories of children to assess their growth with respect to their peers, as well as to anticipate future growth and disease risk. While retrospective modelling of BMI trajectories has been an active area of research, prospective prediction of continuous BMI trajectories from historical growth data has not been well investigated. Using weig…
▽ More
Clinicians often investigate the body mass index (BMI) trajectories of children to assess their growth with respect to their peers, as well as to anticipate future growth and disease risk. While retrospective modelling of BMI trajectories has been an active area of research, prospective prediction of continuous BMI trajectories from historical growth data has not been well investigated. Using weight and height measurements from birth to age 10 years from a longitudinal mother-offspring cohort, we leveraged a multi-task Gaussian processes model, called MagmaClust, to derive probabilistic predictions for BMI trajectories over various forecasting periods. Experiments were conducted to evaluate the accuracy, sensitivity to missing values, and number of clusters. The results were compared with cubic B-spline regression and a parametric Jenss-Bayley mixed effects model. A downstream tool computing individual overweight probabilities was also proposed and evaluated. In all experiments, MagmaClust outperformed conventional models in prediction accuracy while correctly calibrating uncertainty regardless of the missing data amount (up to 90\% missing) or the forecasting period (from 2 to 8 years in the future). Moreover, the overweight probabilities computed from MagmaClust's uncertainty quantification exhibited high specificity ($0.94$ to $0.96$) and accuracy ($0.86$ to $0.94$) in predicting the 10-year overweight status even from age 2 years. MagmaClust provides a probabilistic non-parametric framework to prospectively predict BMI trajectories, which is robust to missing values and outperforms conventional BMI trajectory modelling approaches. It also clusters individuals to identify typical BMI patterns (early peak, adiposity rebounds) during childhood. Overall, we demonstrated its potential to anticipate BMI evolution throughout childhood, allowing clinicians to implement prevention strategies.
△ Less
Submitted 4 February, 2024;
originally announced February 2024.
-
Bayesian Nonparametric Vector Autoregressive Models via a Logit Stick-breaking Prior: an Application to Child Obesity
Authors:
Mario Beraha,
Alessandra Guglielmi,
Fernando A. Quintana,
Maria de Iorio,
Johan Gunnar Eriksson,
Fabian Yap
Abstract:
Overweight and obesity in adults are known to be associated with risks of metabolic and cardiovascular diseases. Because obesity is an epidemic, increasingly affecting children, it is important to understand if this condition persists from early life to childhood and if different patterns of obesity growth can be detected. Our motivation starts from a study of obesity over time in children from So…
▽ More
Overweight and obesity in adults are known to be associated with risks of metabolic and cardiovascular diseases. Because obesity is an epidemic, increasingly affecting children, it is important to understand if this condition persists from early life to childhood and if different patterns of obesity growth can be detected. Our motivation starts from a study of obesity over time in children from South Eastern Asia. Our main focus is on clustering obesity patterns after adjusting for the effect of baseline information. Specifically, we consider a joint model for height and weight patterns taken every 6 months from birth. We propose a novel model that facilitates clustering by combining a vector autoregressive sampling model with a dependent logit stick-breaking prior. Simulation studies show the superiority of the model to capture patterns, compared to other alternatives. We apply the model to the motivating dataset, and discuss the main features of the detected clusters. We also compare alternative models with ours in terms of predictive performances.
△ Less
Submitted 23 March, 2022;
originally announced March 2022.
-
Unbiased approximation of posteriors via coupled particle Markov chain Monte Carlo
Authors:
Willem van den Boom,
Ajay Jasra,
Maria De Iorio,
Alexandros Beskos,
Johan G. Eriksson
Abstract:
Markov chain Monte Carlo (MCMC) is a powerful methodology for the approximation of posterior distributions. However, the iterative nature of MCMC does not naturally facilitate its use with modern highly parallel computation on HPC and cloud environments. Another concern is the identification of the bias and Monte Carlo error of produced averages. The above have prompted the recent development of f…
▽ More
Markov chain Monte Carlo (MCMC) is a powerful methodology for the approximation of posterior distributions. However, the iterative nature of MCMC does not naturally facilitate its use with modern highly parallel computation on HPC and cloud environments. Another concern is the identification of the bias and Monte Carlo error of produced averages. The above have prompted the recent development of fully ('embarrassingly') parallel unbiased Monte Carlo methodology based on coupling of MCMC algorithms. A caveat is that formulation of effective coupling is typically not trivial and requires model-specific technical effort. We propose coupling of MCMC chains deriving from sequential Monte Carlo (SMC) by considering adaptive SMC methods in combination with recent advances in unbiased estimation for state-space models. Coupling is then achieved at the SMC level and is, in principle, not problem-specific. The resulting methodology enjoys desirable theoretical properties. A central motivation is to extend unbiased MCMC to more challenging targets compared to the ones typically considered in the relevant literature. We illustrate the effectiveness of the algorithm via application to two complex statistical models: (i) horseshoe regression; (ii) Gaussian graphical models.
△ Less
Submitted 27 April, 2023; v1 submitted 8 March, 2021;
originally announced March 2021.
