Importance & Interpretation of Feature Importance in Time-Series Models

Sowmya Ramesh Kumar

Abstract: Feature importance in time series models is a critical component of data science, playing a substantial role in interpreting variable contributions and augmenting the overall comprehension of model behavior. The analysis of time series data, characterized by its temporal ordering, presents unique challenges and considerations when it comes to feature importance. As data becomes noisier and challenging to predict, we adapt to sophisticated models like Machine Learning (ML) or Deep Learning (DL) to get accurate predictions. We know that ML/ DL Models are a black box and harder to explain. In this extensive exploration, we will delve into the concept of feature importance, its relevance in time series models, and various methods for interpreting variable contributions.

Keywords: Time - series, feature importance, multi - variate models

How to Cite?: Sowmya Ramesh Kumar, "Importance & Interpretation of Feature Importance in Time-Series Models", Volume 7 Issue 2, February 2018, International Journal of Science and Research (IJSR), Pages: 1667-1669, https://www.ijsr.net/getabstract.php?paperid=SR24213012103, DOI: https://dx.doi.org/10.21275/SR24213012103