Deep Generalised Mixed Models: a Novel Neural Network Structure for Analysing Hierarchical Data

arXiv:2608.05930v1 Announce Type: cross Abstract: The experience sampling method (ESM) is a longitudinal research design where participants report their thoughts, emotional states and behaviours multiple times a day. Our work is motivated by such data collected by the GrowIt! app, which was released to investigate daily emotions among adolescents during the COVID-19 pandemic. Current procedures to analyse ESM data face various challenges. While standard statistical techniques may not scale well to a high-dimensional setting, machine learning procedures can give biased results due to selection bias introduced by missingness. In our motivating dataset, adolescents dropped out due to previous strong feelings of negative emotions. Hence, the implied missing data are of the missing-at-random type that standard machine learning procedures cannot accommodate. We develop a novel neural network architecture that generalises mixed effects models to deep learning to overcome these challenges. It allows semi-parametric and flexible modelling of data's mean and correlation structure through fixed and random effects. For estimation, we use an adaptation of variational auto-encoders and a Bayesian data augmentation algorithm. Through this approach, the model can accommodate longitudinal outcomes following generic distributions, scale well to high-dimensional settings and provide valid inference when data are missing-at-random. We applied the Deep Generalised Mixed Model to the GrowIt! study and various simulations. The results show potential for the Deep Generalised Mixed Model, yet suboptimal performance due to model instability.

Important: e-prints posted on arXiv are not peer-reviewed by arXiv; they should not be relied upon without context to guide clinical practice or health-related behavior and should not be reported in news media as established information without consulting multiple experts in the field.

Abstract:The experience sampling method (ESM) is a longitudinal research design where participants report their thoughts, emotional states and behaviours multiple times a day. Our work is motivated by such data collected by the GrowIt! app, which was released to investigate daily emotions among adolescents during the COVID-19 pandemic. Current procedures to analyse ESM data face various challenges. While standard statistical techniques may not scale well to a high-dimensional setting, machine learning procedures can give biased results due to selection bias introduced by missingness. In our motivating dataset, adolescents dropped out due to previous strong feelings of negative emotions. Hence, the implied missing data are of the missing-at-random type that standard machine learning procedures cannot accommodate. We develop a novel neural network architecture that generalises mixed effects models to deep learning to overcome these challenges. It allows semi-parametric and flexible modelling of data's mean and correlation structure through fixed and random effects. For estimation, we use an adaptation of variational auto-encoders and a Bayesian data augmentation algorithm. Through this approach, the model can accommodate longitudinal outcomes following generic distributions, scale well to high-dimensional settings and provide valid inference when data are missing-at-random. We applied the Deep Generalised Mixed Model to the GrowIt! study and various simulations. The results show potential for the Deep Generalised Mixed Model, yet suboptimal performance due to model instability.

Machine Learning (stat.ML); Machine Learning (cs.LG)

arXiv:2608.05930 [stat.ML]

(or arXiv:2608.05930v1 [stat.ML] for this version)

https://doi.org/10.48550/arXiv.2608.05930 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

figures/S2A_10_M-eps-converted-to.pdf

figures/S2A_10_M.eps

figures/S2A_10_M_b-eps-converted-to.pdf

figures/S2A_10_M_b.eps

figures/S2A_10_Q-eps-converted-to.pdf

figures/S2A_10_Q.eps

figures/S2A_10_Q_b-eps-converted-to.pdf

figures/S2A_10_Q_b.eps

figures/S2A_30_M-eps-converted-to.pdf

figures/S2A_30_M.eps

figures/S2A_30_M_b-eps-converted-to.pdf

figures/S2A_30_M_b.eps

figures/S2A_30_Q-eps-converted-to.pdf

figures/S2A_30_Q.eps

figures/S2A_30_Q_b-eps-converted-to.pdf

figures/S2A_30_Q_b.eps

figures/S2B_10_M-eps-converted-to.pdf

figures/S2B_10_M.eps

figures/S2B_10_M_b-eps-converted-to.pdf

figures/S2B_10_M_b.eps

figures/S2B_10_Q-eps-converted-to.pdf

figures/S2B_10_Q.eps

figures/S2B_10_Q_b-eps-converted-to.pdf

figures/S2B_10_Q_b.eps

figures/S2B_30_M-eps-converted-to.pdf

figures/S2B_30_M.eps

figures/S2B_30_M_b-eps-converted-to.pdf

figures/S2B_30_M_b.eps

figures/S2B_30_Q-eps-converted-to.pdf

figures/S2B_30_Q.eps

figures/S2B_30_Q_b-eps-converted-to.pdf

figures/S2B_30_Q_b.eps

(31 additional files not shown) You must enabled JavaScript to view entire file list.

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

aiqbench analysis

Deep Generalised Mixed Models: a Novel Neural Network Structure for Analysing Hierarchical Data

Related coverage