Abstract:
The article substantiates and proves the expediency of using economic-mathematical
modeling for the formation of a forecast of economic trends and identification of probable
ways of development of socio-economic phenomena and processes. These circumstances
determine the relevance of in-depth research into the process of forecasting
using mathematical methods and evaluation of the adopted decision.
The purpose of the work is the use of modern tools of analytical and simulation economic-
mathematical modeling for forecasting the development trends of economic entities
in conditions of uncertainty.
An analysis of methods and models for forecasting time series and determination of the
most effective combinations of them for forecasting economic phenomena and processes
was carried out, and the possibility of using them in practice for the analysis and
planning of the activities of economic entities was investigated.
The option of solving the problem of forecasting economic development trends was
carried out on the basis of statistical data, using the example of hotel business enterprises.
Methods and models of time series research and forecasting were used in the
work: correlation analysis, autoregression and moving average methods, artificial neural
network (ANN) models, and autoregressive moving average (ARIMA) model. The results
showed that both the ARIMA model and the ANN model can be effectively used for
forecasting tasks. It is proven that the ANN model has a higher prediction accuracy at
time intervals that are close to the original data. At the same time, the ARIMA model is
more appropriate for long-term forecasting. The obtained results allow us to put forward
ideas about the simultaneous use of both models, which can compensate for the shortcomings
of each of them. Also, the models can be used separately for more accurate
forecasting of values for the required time period. More effective is the method by which
artificial neural networks can be applied to solve the problem of clustering. This will
allow you to single out ranges for forecasting. And then apply ARIMA forecasting to the
obtained data sets. The proposed algorithm can be used to determine trends in the
development of the hotel industry, as its application reduces the risk of forecasting
errors.
The results of the work consist of practical recommendations regarding the features of
the application of economic and mathematical modeling methods for the construction
of forecast indicators and prospects for the development of economic entities. The built
model uses the properties of basic forecasting models, which allows for an increase in
the degree of reliability and validity of scientific research.
