The missing value estimation in numerical modeling of complex dynamic systems

This paper is deals with the problem of missing values in the modeling of complex dynamic systems. Considered different types of missing values and general approaches to dealing with missing data. This paper provides an overview of the most common methods of missing data estimation, and presents an original hybrid adaptive method of estimation with neuro-fuzzy control. Evaluate the effectiveness of various methods of missing data estimating applied to data sets that contain information about the processes in complex dynamic systems. Special attention is paid to the impact of missing data presence on the effectiveness of the models.