CRAN version


Learning Bayesian Networks with Mixed Variables

The deal package offers R functions that estimate Bayesian networks with continuous and/or discrete variables can be learned and compared from data. It includes several methods for analysing data using Bayesian networks with variables of discrete and/or continuous types but restricted to conditionally Gaussian networks. Construction of priors for network parameters is supported and their parameters can be learned from data using conjugate updating. The network score is used as a metric to learn the structure of the network and forms the basis of a heuristic search strategy. deal has an interface to Hugin. The method is described in further detail in Boettcher and Dethlefsen (2003).


The released and tested version of deal is available at the Comprehensive R Archive Network). It is installed from within R by running


To install the latest version of deal directly from GitHub, run


Ensure that you have the package development prerequisites if you wish to install the package from the source. For previous versions of deal, visit the archive at CRAN.


  1. Boettcher, S., & Dethlefsen, C. (2003). deal: A Package for Learning Bayesian Networks. Journal of Statistical Software, 8(20), 1 - 40. doi: