Skip to main content

Table 1 Patterns of missing data

From: How to deal with missing longitudinal data in cost of illness analysis in Alzheimer’s disease—suggestions from the GERAS observational study

Missing data pattern Description
Missing completely at random (MCAR) • Data are missing for reasons not related to observed or unobserved variables
• Simple statistical approaches to deal with missing data can provide unbiased results
Missing at random (MAR) • Probability of missingness is related to observed data but not to unobserved data
• MAR is the assumption for most imputation methods
Missing not at random (MNAR) • Missingness is related to unobserved data
  1. It is not possible to distinguish between MAR and MNAR based on observed data