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, Missing At Random, and Time-dependent Missing At Random data have less bias and variability than 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 either default package settings or a lag-1 imputation model, multiple imputation struggled to recover the parameters.

翻译：在生态瞬时评估数据（EMA）的应用中，缺失数据普遍存在，因为参与者流失是常见问题。因此，任何EMA研究都必须制定缺失数据处理方案。本文探讨了时间序列分析中的缺失现象，以及将数据建模为离散时间连续测量状态空间模型时处理缺失数据的适当方法。研究发现，完全随机缺失、随机缺失和时间依赖随机缺失数据相较于自回归时间依赖随机缺失和非随机缺失数据具有更低的偏差和变异性。卡尔曼滤波在处理缺失数据方面表现优异。与现有文献结论相反，我们发现无论采用默认软件包设置还是滞后一阶插值模型，多重插值法均难以有效恢复参数。