Searching for Approximate Solutions in Statistical Modeling Complexity Reduction
Abstract
Purpose: Analysis of the available methods of statistical modeling complexity reduction shows how important it is to efficiently buiild a simplified model of the system in question. The goal of the paper is developing an approach to find an approximate solution of the system which would allow us to expand the applicability of statistical modeling complexity reduction methods. Results: The importance of searching for new methods of building simplified systems has been grounded. An approach to simplified modeling has been developed. To find a simplified solution of the system, it is proposed to use a heightmap which reduces the modeling laboriousness. This approach is highly efficient, with no need of prior exhaustive knowledge about the statistic properties of the system. The method has been shown to be highly efficient for systems which are difficult to describe by polynomial models. Practical relevance: The obtained results can be used for optimizing the efficiency of statistical modeling complexity reduction when the statistical characteristics of the modelled systems are unknown.Published
2015-02-20
How to Cite
Emeljanov, V., & Dokuchaeva, A. (2015). Searching for Approximate Solutions in Statistical Modeling Complexity Reduction. Information and Control Systems, (1), 43-49. https://doi.org/10.15217/issn1684-8853.2015.1.43
Issue
Section
System and process modeling