Shiny app for demonstration of the effect of various imputation methods on data with different missing data mechanisms. It aims to provide an interactive platform and encapsulate the material from the Incomplete Data Analysis course at The University of Edinburgh (delivered in 2019 by Vanda Inácio and Simon Taylor).
The simulation study considers data that are:
- missing completely at random (MCAR)
- missing at random (MAR)
- missing not at random (MNAR)
These are drawn from a bivariate normal distribution and the missing data pattern is univariate (for visualisation purposes).
The app can be accessed here.