Time series analysis, prediction methods, linear methods, mathematical statistics, applied statistics.
Basic methods of quantitative forecasting, causal models, time series. Model performance evaluation, describing statistics, MAE, MAPE, RMSE, entropy measures, naive models. Basic theory of the linear prediction models, covariance and correlation coefficients, smoothing methods, regression methods, Box-Jenkins methodology, statistical tests, genetics algorithms.
Objective of the course is to introduce students to the problems of time series forecasting with a focus on different methods of prediction and evaluation of its quality. Emphasis is also placed on finding solutions to the various methods and models using concrete examples from practice.