Daily dynamics and weekly rhythms: A tutorial on seasonal autoregressive-moving average models combined with...

Area of Science:

• Psychological research methodology
• Time series analysis
• Quantitative psychology

Background:

• Daily emotional experiences are typically modeled using first-order autoregressive models.
• These models often overlook weekly rhythms and delayed week-to-week dynamics in emotional experiences.
• Day-of-the-week effects (DOWEs) and week-to-week dynamics are largely ignored in current psychological research.

Purpose of the Study:

• To introduce visualization techniques for detecting weekly rhythms and day-to-day dynamics in time series data.
• To present and extend seasonal autoregressive-moving average (SARMA) models with DOWEs for analyzing psychological data.
• To encourage researchers to consider dynamics beyond lag-1 autoregressive modeling.

Main Methods:

• Developed complementary visualization techniques for time series data.
• Introduced and extended SARMA models with DOWEs from econometrics.
• Provided an R tutorial for fitting these models, including model fit and selection.
• Applied the models to a daily diary dataset from 98 individuals.

Main Results:

• Identified weekly rhythms and day-to-day dynamics in emotional experiences.
• Demonstrated that most individuals show patterns not captured by current standard practices.
• Highlighted the presence of week-to-week dynamics influencing emotional experiences.

Conclusions:

• Current psychological research practices may inadequately capture the complexity of emotional dynamics.
• There is a need to move beyond the ubiquitous lag-1 autoregressive modeling paradigm.
• Future research should consider dynamics across different timescales for a more comprehensive understanding of emotional experiences.