A Comprehensive Survey of Regression Based Loss Functions for Time Series Forecasting

TL;DR

This survey reviews 14 regression loss functions used in time series forecasting, highlighting their applications, performance across datasets, and providing guidance for selecting baseline objectives.

Contribution

It systematically summarizes and compares diverse loss functions for time series forecasting, offering insights into their suitability and performance.

Findings

Certain loss functions perform consistently well across datasets.

Categories of loss functions can serve as reliable baselines.

Guidelines for choosing loss functions based on data characteristics.

Abstract

Time Series Forecasting has been an active area of research due to its many applications ranging from network usage prediction, resource allocation, anomaly detection, and predictive maintenance. Numerous publications published in the last five years have proposed diverse sets of objective loss functions to address cases such as biased data, long-term forecasting, multicollinear features, etc. In this paper, we have summarized 14 well-known regression loss functions commonly used for time series forecasting and listed out the circumstances where their application can aid in faster and better model convergence. We have also demonstrated how certain categories of loss functions perform well across all data sets and can be considered as a baseline objective function in circumstances where the distribution of the data is unknown. Our code is available at GitHub: https://github.com/aryan-jadon/Regression-Loss-Functions-in-Time-Series-Forecasting-Tensorflow.