The book is devoted to models of statistical multi-step and Markov decision-making processes in conditions of information deficit. Properties of sensitivity, stability, stability, regularity and marginality of Bayes solutions are considered, as well as principles of construction and use of uncertainty and inaccuracy functions.