Comparative Study on Exponentially Weighted Moving Average Approaches for the Self-Starting Forecasting

Abstract

Recently, a number of data analysists have suffered from an insufficiency of historical observations in many real situations. To address the insufficiency of historical observations, self-starting forecasting process can be used. A self-starting forecasting process continuously updates the base models as new observations are newly recorded, and it helps to cope with inaccurate prediction caused by the insufficiency of historical observations. This study compared the properties of several exponentially weighted moving average methods as base models for the self-starting forecasting process. Exponentially weighted moving average methods are the most widely used forecasting techniques because of their superior performance as well as computational efficiency. In this study, we compared the performance of a self-starting forecasting process using different existing exponentially weighted moving average methods under various simulation scenarios and real case datasets. Through this study, we can provide the guideline for determining which exponentially weighted moving average method works best for the self-starting forecasting process.

Ecosystem Photosynthesis in Land‐Surface Models: A First‐Principles Approach Incorporating Acclimation

“…We therefore tested whether the longer‐term acclimation timescales could be mimicked based on a technique used to incorporate memory in other aspects of climate modeling, the exponential weighted moving average method. This method, here called the weighted mean approach, is used in forecasting systems that deal with inaccurate prediction caused by the insufficiency of historical observations and allows for a self‐starting forecasting process without having to store past data (Yu et al., 2020). The method is used in a variety of applications in forecasting, from estimating soil moisture from precipitation (Campos de Oliveira et al., 2017) to vegetation acclimation processes (e.g., Vanderwel et al., 2015).…”

A Reversible Automatic Selection Normalization (RASN) Deep Network for Predicting in the Smart Agriculture System

“…The model parameters have been fixed in the testing phase, and the fundamental values such as mean and variance are matched. The exponentially weighted average method [50] is used to record the mean and variance of each batch so that the values used in the testing phase are close to the distribution of the total sample, and the calculation formulae are shown in Equations ( 5)- (9).…”

Big Data and AI for Process Innovation in the Industry 4.0 Era

“…Dualkernel-based aggregated residual networks were proposed for a visual inspection system's automatic detection of injection mold process defects. Another neural-network-based approach that was presented in the "Comparative Study on Exponentially Weighted Moving Average Approaches for the Self-Starting Forecasting" by Yu, Kim, Bai, and Han [12] focused on applications of exponentially weighted moving average (EWMA) models using time-series information as a forecasting process. After conducting simulation scenarios, they recommended using a two-stage EWMA model as a base model for conducting a self-stating forecasting process for complex time series.…”