Missing Data in Discrete Time State-Space Modeling of Ecological Momentary Assessment Data: A Monte-Carlo Study of Imputation Methods

When using ecological momentary assessment data (EMA), missing data is pervasive as participant attrition is a common issue. Thus, any EMA study must have a missing data plan. In this paper, we discuss missingness in time series analysis and the appropriate way to handle missing data when the data is modeled as a discrete time continuous measure state-space model. We found that Missing Completely At Random and Time-dependent Missing At Random data have less bias and variability than Missing At Random, Autoregressive Time-dependent Missing At Random, and Missing Not At Random. The Kalman filter excelled at handling missing data. Contrary to the literature, we found that, with default package settings, multiple imputation struggled to recover the parameters.