A Novel Hybrid Approach for Time Series Forecasting: Period Estimation and Climate Data Analysis Using Unsupervised Learning and Spline Interpolation

TL;DR

This paper presents a hybrid method for time series forecasting of climate data, combining period estimation via unsupervised learning and spline interpolation with ensemble modeling to improve forecast accuracy.

Contribution

It introduces a new algorithm for period detection using unsupervised learning and spline interpolation, enhancing climate data analysis and forecasting accuracy.

Findings

Improved forecast accuracy through ensemble modeling.

Effective period estimation using unsupervised learning.

Enhanced climate data analysis techniques.

Abstract

This article explores a novel approach to time series forecasting applied to the context of Chennai's climate data. Our methodology comprises two distinct established time series models, leveraging their strengths in handling seasonality and periods. Notably, a new algorithm is developed to compute the period of the time series using unsupervised machine learning and spline interpolation techniques. Through a meticulous ensembling process that combines these two models, we achieve optimized forecasts. This research contributes to advancing forecasting techniques and offers valuable insights into climate data analysis.