vector autoregressive model
Definition
Vector autoregressive model is a variation of multiple regression where independent variables are lagged forms of the dependent variables, capturing temporal dynamics between network elements through lagged relationships—the predicted relationships between a variable at one time point and variables at the next time point. In idiographic psychological networks, VAR models illustrate how variables such as symptoms or emotions are associated over time, with dependent variables functioning as a combination of their own lagged values and the lagged values of all other variables in the system. The fixed moderated time series model extends this approach as a VAR-based model in which all parameters can be moderated, including the innovation structure, while also directly estimating changes in mean levels of variables through a state space framework.
Sources: Bringmann et al. (2024)
Related Terms
Applications
Vector Autoregressive Model and Idiographic Networks
Idiographic psychological networks based on intensive longitudinal data employ vector autoregressive models as their underlying framework to study person-specific associations among psychological variables over time. The VAR model captures temporal dynamics through lagged relationships, allowing researchers to identify how symptoms or emotions influence each other in an individual's psychological system.
Sources: Bringmann et al. (2024)
Vector Autoregressive Model and Moderation Analysis
The fixed moderated time series model addresses the application of moderation to idiographic VAR-based networks by allowing all parameters of a VAR model—including the innovation structure—to be moderated by contextual factors. This enables researchers to investigate how external variables influence the temporal dynamics.
Sources: Bringmann et al. (2024)
Vector Autoregressive Model and Intensive Longitudinal Data
Vector autoregressive models are applied to intensive longitudinal data to model temporal dynamics in psychological variables. VAR-based approaches can capture both the associations among variables and changes in these associations over time.
Sources: Bringmann et al. (2024)
