Linear, Machine Learning and Probabilistic Approaches for Time Series Analysis

TL;DR

This paper compares linear, machine learning, and probabilistic methods for time series analysis, highlighting their applications and performance in forecasting and probabilistic modeling.

Contribution

It provides a comprehensive overview of various modeling approaches, including linear, machine learning, and probabilistic methods, with experimental results.

Findings

Different model combinations show varying forecasting accuracy.

Probabilistic approaches using copulas and Bayesian inference are effective.

Machine learning models outperform traditional linear models in some cases.

Abstract

In this paper we study different approaches for time series modeling. The forecasting approaches using linear models, ARIMA alpgorithm, XGBoost machine learning algorithm are described. Results of different model combinations are shown. For probabilistic modeling the approaches using copulas and Bayesian inference are considered.