Economic trends forecasting in the development of hotel business enterprises

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.