INTELLECTUAL TECHNOLOGIES AND DECISION SUPPORT SYSTEMS FOR THE CONTROL OF THE ECONOMIC AND FINANCIAL PROCESSES

摘要

A computer based decision support system is proposed the basic tasks of which are adaptive model constructing and forecasting of various types of processes that are developing in socio-economic systems under the influence of fundamental structural changes. The complexity and urgency of the solvable problem is the need to provide acceptable quality forecasts of financial and economic indicators for short data samples, when the usage of retrospective data is impossible or significantly limited. The DSS development is based on the system analysis principles, i.e. the possibility for taking into consideration of some stochastic and information uncertainties, forming alternatives for models and forecasts, and tracking of the computing procedures correctness during all stages of data processing. A modular architecture is implemented that provides a possibility for the further enhancement and modification of the system functional possibilities with new forecasting and parameter estimation techniques. In addition, the proposed system, thanks to the modular architecture, can be improved by using the software of different vendors without any additional structural changes. A high quality of the final result is achieved thanks to appropriate tracking of the computing procedures at all stages of data processing during computational experiments: preliminary data processing, model constructing, and forecasts estimation. The tracking is performed with appropriate sets of statistical quality parameters. Example is given for estimation of financial risk in insurance sphere and the electricity consumption in terms of energy saving. The examples solved show that the system developed has good perspectives for the practical use. It is supposed that the system will be universal and find its applications as an extra tool for support of decision making when developing the strategies for companies and enterprises of various types.