Theoretical examination of Bayesian time series and an example application

Time series analyzes are widely used in forecasting for the future. The use of Bayesian time series analysis is increasing day by day with the development of simulation methods and the spread of computer-based programs. In this study, the estimation of the electric car sales figures for the future period with the Bayesian time series has been examined. For this purpose, Bayesian time series analysis was carried out using the data of electric car sales figures for the 72-month period between January 2016 and December 2021, and electric car sales figures for 2023 were estimated. In addition, in order to have information about the state of the automobile market, the estimation of automobile sales in the general market and the estimation of what percentage of the electric cars in the general market are made using the bayesian time series. In order to understand the success of all these estimations, the estimations made by the Bayesian method were compared with the estimations made by the classical method. For the Bayesian time series analysis of the study, the appropriate code structure was created and the R program was used. Time series analysis was carried out in the EViews program. In order to determine the success of the estimation results, the square root values of the mean squared error were checked and the results that gave the smallest value were determined as the results obtained as a result of the Bayesian time series analysis. It is very important to use forecasts with minimum error when making forecasts for the future, both for the success of forecast results and for planning for the future. With this study, it has been revealed that Bayesian time series analysis is a suitable and effective method for predicting the future.