Modeling Nonstationary Emotion Dynamics in Dyads using a Time-Varying Vector-Autoregressive Model

Abstract

Emotion dynamics are likely to arise in an interpersonal context. Standard methods to study emotions in interpersonal interaction are limited because stationarity is assumed. This means that the dynamics, for example, time-lagged relations, are invariant across time periods. However, this is generally an unrealistic assumption. Whether caused by an external (e.g., divorce) or an internal (e.g., rumination) event, emotion dynamics are prone to change. The semi-parametric time-varying vector-autoregressive (TV-VAR) model is based on well-studied generalized additive models, implemented in the software R. The TV-VAR can explicitly model changes in temporal dependency without pre-existing knowledge about the nature of change. A simulation study is presented, showing that the TV-VAR model is superior to the standard time-invariant VAR model when the dynamics change over time. The TV-VAR model is applied to empirical data on daily feelings of positive affect (PA) from a single couple. Our analyses indicate reliable changes in the male's emotion dynamics over time, but not in the female's-which were not predicted by her own affect or that of her partner. This application illustrates the usefulness of using a TV-VAR model to detect changes in the dynamics in a system.

A Tutorial on Estimating Time-Varying Vector Autoregressive Models

“…In this section, we explain how a time-varying VAR model can be estimated using the Generalized Additive Model (GAM) framework (see also Bringmann et al, 2018. GAMs are extensions of linear models which allow to represent highly nonlinear functions by adding a numbers of basis functions.…”

“…Different model specifications of the time-varying model, and stationary models can be compared either by using information criteria such as the AIC (see e.g., Bringmann et al, 2018), or by using data driven approaches. For example, one could divide the time series into a training and test set using some time-stratification to ensure that both data sets equally represent the entire period of the time series.…”

Foundations of idiographic methods in psychology and applications for psychotherapy

“…However, TV-AR is limited in that it can only model a single variable over time. A second technique, time-varying vector autoregression (TV-VAR), is a more flexible method for studying a time-varying multivariate time series (Bringmann, Ferrer, Hamaker, Borsboom, & Tuerlinckx, 2018;Haslbeck, Bringmann, & Waldorp, 2017). However, time-varying methods often require a large number of observations (based on experience and discussion with researchers, our best guess is that over 100 time points may often be needed; L.F. Bringmann, personal communication, July 6, 2018).…”

A Clinician’s Primer for Idiographic Research: Considerations and Recommendations

“…However, not all idiographic methods can test different lags within the same model and those that do require a considerable number of observations (e.g., Bringmann, Ferrer, Hamaker, Borsboom, & Tuerlinckx, 2018, Bringmann, L.F., personal correspondence, July 6, 2018. Thus, researchers have tended to construct models at lag 1 , most likely due to power constraints.…”