This book uses the EM (expectation maximization) algorithm to simultaneously estimate the missing data & unknown parameter (s) associated with a data set. The parameters describe the component distributions of the mixture; the distributions may be continuous or discrete.