A graphical vector autoregressive modelling approach to the analysis of electronic diary data

Abstract

BackgroundIn recent years, electronic diaries are increasingly used in medical research and practice to investigate patients' processes and fluctuations in symptoms over time. To model dynamic dependence structures and feedback mechanisms between symptom-relevant variables, a multivariate time series method has to be applied.MethodsWe propose to analyse the temporal interrelationships among the variables by a structural modelling approach based on graphical vector autoregressive (VAR) models. We give a comprehensive description of the underlying concepts and explain how the dependence structure can be recovered from electronic diary data by a search over suitable constrained (graphical) VAR models.ResultsThe graphical VAR approach is applied to the electronic diary data of 35 obese patients with and without binge eating disorder (BED). The dynamic relationships for the two subgroups between eating behaviour, depression, anxiety and eating control are visualized in two path diagrams. Results show that the two subgroups of obese patients with and without BED are distinguishable by the temporal patterns which influence their respective eating behaviours.ConclusionThe use of the graphical VAR approach for the analysis of electronic diary data leads to a deeper insight into patient's dynamics and dependence structures. An increasing use of this modelling approach could lead to a better understanding of complex psychological and physiological mechanisms in different areas of medical care and research.

Longitudinal and personalized networks of eating disorder cognitions and behaviors: Targets for precision intervention a proof of concept study

“…Next, we used the graphical VAR package in R to create intraindividual networks for three randomly selected individuals with AN (Wild et al, ). We created both temporal and contemporaneous networks for each of these individuals.…”

“…In all three networks, each symptom (or variable) is represented as a node, and links between nodes (i.e., edges) represent associations between symptoms. In a temporal network, edges with an arrowhead pointing from one node to another indicates that the first node predicts the other at the next Next, we used the graphical VAR package in R to create intraindividual networks for three randomly selected individuals with AN (Wild et al, 2010). We created both temporal and contemporaneous networks for each of these individuals.…”

The Gaussian Graphical Model in Cross-Sectional and Time-Series Data

“…In addition to temporal effects, VAR analyses also include contemporaneous effects, which can be modeled as a GGM. We will term this modeling framework (a VAR model with contemporaneous effects explicitly modeled and portrayed as a GGM) graphical VAR (GVAR; Wild et al, 2010). 13 A useful equivalent way to denote a GVAR model is by using a conditional Gaussian distribution:…”

“…Indeed, Equation (1) makes this assumption and all other equations follow from this. The assumption of normality is not without 18 We standardized every dataset before analyzing and used the standardization of Wild et al (2010) for temporal networks in n = 1 and pooled temporal networks. GGMs are readily standardized by using partial correlation coefficients (Equation (2)), which have been used in all GGMs shown in this paper.…”

How are you feeling?: A personalized methodology for predicting mental states from temporally observable physical and behavioral information

“…Vector Autoregression (VAR) [22] models, including regression analyses, have been successfully used to capture linear interdependencies among multiple univariate time series, and have been shown effective in forecasting tasks in financial [23, 24], meteorology [25, 26], biomedical [27, 28] domains, etc.…”