Multilevel Autoregressive Models when the Number of Time Points is Small

Gistelinck, Fien; Loeys, Tom; Flamant, Nele

doi:10.1080/10705511.2020.1753517

Cited by 15 publications

(14 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As indicated earlier, the first measurement point is likely to be influenced by the previous time points which cannot be modelled, and the effects of time-invariant confounders on the first measurement reflect these accumulating effects over time. As such, we cannot treat the first measurement point in the same way as the following time points (Gistelinck et al, 2021).…”

Section: Figure 4 a Path Diagram Of Dynamic Panel Model (Dpm)mentioning

confidence: 99%

Thinking clearly about time-invariant confounders in cross-lagged panel models: A guide for choosing a statistical model from a causal inference perspective

Murayama¹,

Gfrörer²

2022

Preprint

View full text Add to dashboard Cite

Many models have been proposed to examine the reciprocal cross-lagged causal effects from panel data. The purpose of the current article is to clarify how these various models control for unmeasured time-invariant confounders, helping researchers understand the differences of the models from causal inference perspective. We showed that cross-lagged panel model controls for time-invariant confounders that indirectly influence the relationships from prior time points. We also showed that there are other models (e.g., dynamic panel model, random-intercept cross-lagged panel model) that directly control for time-invariant confounders using time-invariant latent factor(s). These models vary in terms of the type of time-invariant confounders that are controlled as well whether the model assumes simple or complex individual trajectories over time. Based on the comparisons, we made some practical suggestions for researchers to select a model. We close the discussion by noting that those models that directly incorporate unmeasured time-invariant confounders rest on some critical and often unrealistic assumptions, providing a caution to over-trust causal estimates.

show abstract

Section: Figure 4 a Path Diagram Of Dynamic Panel Model (Dpm)mentioning

confidence: 99%

Thinking clearly about time-invariant confounders in cross-lagged panel models: A guide for choosing a statistical model from a causal inference perspective

Murayama¹,

Gfrörer²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The first is Lüdtke’s bias, which involves observed means being potentially unreliable and a potentially inaccurate quantification of the true trait-like component of the variable (Lüdtke et al, 2008). Differences between latent centering and observed centering can be minimal in standard mixed effect models, but the choice of centering has a non-negligible effect on carryover effect estimates (i.e., φm and φy in models for intensive longitudinal data; Gistelinck et al, 2021). This is captured by the second type of bias—Nickell’s bias (Nickell, 1981)—whereby carryover effects are underestimated when centering around an observed person-mean.…”

Section: Challenges With Intensive Longitudinal Datamentioning

confidence: 99%

“…Rather than centering around the observed means to disaggregate variables into within- and between-person components, DSEM centers around a latent mean that accounts for unreliability and measurement error. Latent centering has been shown to eliminate both Ludtke’s bias and Nickell’s bias in multilevel time-series models with lagged predictors (Asparouhov et al, 2018) so long at the number of repeated measures is about 10 or more (Gistelinck et al, 2021). Also similar to structural equation modeling, DSEM can naturally accommodate multivariate models that have multiple outcomes, meaning that no special data preprocessing (like stacking) is needed.…”

Section: Longitudinal Mediation As a Dynamic Structural Equation Mode...mentioning

confidence: 99%

Intensive longitudinal mediation in Mplus.

McNeish¹,

MacKinnon²

2022

Psychological Methods

View full text Add to dashboard Cite

Much of the existing longitudinal mediation literature focuses on panel data where relatively few repeated measures are collected over a relatively broad timespan. However, technological advances in data collection (e.g., smartphones, wearables) have led to a proliferation of short duration, densely collected longitudinal data in behavioral research. These intensive longitudinal data differ in structure and focus relative to traditionally collected panel data. As a result, existing methodological resources do not necessarily extend to nuances present in the recent influx of intensive longitudinal data and designs. In this tutorial, we first cover potential limitations of traditional longitudinal mediation models to accommodate unique characteristics of intensive longitudinal data. Then, we discuss how recently developed dynamic structural equation models (DSEM) may be well-suited for mediation modeling with intensive longitudinal data and can overcome some of the limitations associated with traditional approaches. We describe four increasingly complex intensive longitudinal mediation models: (1) stationary models where the indirect effect is constant over time and people, (2) person-specific models where the indirect effect varies across people, (3) dynamic models where the indirect effect varies across time, and (4) cross-classified models where the indirect effect varies across both time and people. We apply each model to a running example featuring a mobile health intervention designed to improve health behavior of individuals with binge eating disorder. In each example, we provide annotated Mplus code and interpretation of the output to guide empirical researchers through mediation modeling with this increasingly popular type of longitudinal data.

show abstract

“…Even though multilevel autoregressive model considers the effect of previous data points on the later data points to model the process, the first-and second-order models (i.e. considering only one or two lags) have usually been used in the studies (Gistelinck et al, 2020;H. Goldstein et al, 1994).…”

Section: Established Modeling Strategies To Inform Decision Rules For a Jitaimentioning

confidence: 99%

“…One should note that at this point, there has been some attempts to extend existing approaches in order to deal with these delayed effects. An example of these is the use of lags in multilevel models (Gistelinck et al, 2020;H. Goldstein et al, 1994;Hamaker and Grasman, 2015).…”

Section: Introductionmentioning

confidence: 99%

Dynamical Outcome Modeling of Fragmented Intensive Longitudinal Data

Lagoa¹,

Hojjatinia²,

Conroy³

2021

Preprint

View full text Add to dashboard Cite

The advent of new sensing technologies has enabled the collection of high frequency individual data. Not only can this allow for the identification of personalized behavior models, it also opens up the possibility of developing just-in-time interventions that leverage the information collected to determine when and which micro intervention should be provided. However, there are significant challenges in the analysis of this type of data. First, due to the high rate of data collection, one can no longer assume that the stimulus (independent excitation or micro intervention) only has an instantaneous effect on the outcome, one has to also allow for delayed effects. Moreover, one is also frequently faced with fragmented data; i.e., poor placement of sensors, non-wear of the data collecting device and/or external disturbances can lead to intervals of time where the data collected is not reliable; i.e., missing or corrupted. To deal with these challenges, we leverage concepts from the areas of dynamical systems and signal processing to develop tools that i) can identify models that take into account the delayed stimuli effects and ii) are able to handle fragmented data. In this paper, we provide both the mathematical foundation of the tools proposed and the description of a package that implements them. We also discuss ways to interpret the results obtained.

show abstract

Multilevel Autoregressive Models when the Number of Time Points is Small

Cited by 15 publications

References 28 publications

Thinking clearly about time-invariant confounders in cross-lagged panel models: A guide for choosing a statistical model from a causal inference perspective

Thinking clearly about time-invariant confounders in cross-lagged panel models: A guide for choosing a statistical model from a causal inference perspective

Intensive longitudinal mediation in Mplus.

Dynamical Outcome Modeling of Fragmented Intensive Longitudinal Data

Contact Info

Product

Resources

About