BIMAM - Bayesian IMputation and Analysis Model

The combination of multiple datasets is routine in modern epidemiology, but different studies may have measured different sets of variables; this is often inefficiently dealt with by excluding studies or dropping variables. Multilevel multiple imputation methods to impute these “systematically” missing data (as opposed to “sporadically” missing data within a study) are available, but problems may arise when many random effects are needed in the analysis and imputation models to allow for heterogeneity across studies. BIMAM implements a Bayesian method that works well in this situation.

The Bayesian hierarchical approach implemented in BIMAM simultaneously imputes missing variables across datasets (systematically missing data) and performs the analysis of interest, while allowing to fully account for heterogeneity in both imputation and analysis models. It also imputes any missing values within datasets (sporadically missing data) if present. BIMAM imputes binary and continuous missing data, and analyses binary and continuous outcomes.

The BIMAM tool is a user-friendly, freely available tool that does not require knowledge of Bayesian methods.

You can download the tool with its related documentation by clicking on the icon below, but before you do that, please read these instructions:

  • Before installing BIMAM, please download and install Microsoft MPI (MS-MPI). Version 8.1 or newer is required. This framework allows running parallel applications on the Windows platform.

  • Unzip the downloaded file to extract the exe file

  • Double-click the exe file to start the installation wizard

  • The software should be installed in a folder on which you have writing permissions (e.g. your Documents folder; C:\ProgramFiles may NOT work), as the software writes some files while it is running

  • If, for any reason, you want to install it again, please make sure to uninstall any of its previous installations.