Stationary time series in pricing

Abstract

The methods for analyzing the dynamics of time series are compared in this study, the mechanisms for assessing the accuracy of value forecasting are examined, and a brief description of the models and examples of their use are provided. The problem of choosing the optimal model according to the criterion of the minimum forecasting error is stated and solved. The methods of mathematical modeling, mathematical statistics and econometrics, such as autoregression, moving average, exponential smoothing, and neural network modeling were used to solve this problem. The result of the study is the algorithm for finding the optimal model based on minimizing the forecasting error, as well as the program that implements this algorithm.