Effectiveness of Single and Double Exponential Smoothing: SES, ARRSES and Holt’s Linear for Time Series Data Prediction with Trend and Non-seasonal Characteristic (Covid-19 Vaccinate Case)

Abstract

Purpose of this research is effectiveness the exponential smoothing for predict the time series data which has trend and non-seasonal characteristic. In this research using data case vaccinate of covid-19 1st, 2nd, 3rd and 4th. The advantage of this research is we can choose the best method for this type data. The methodology of this research is quantitative, with analyze data method is exponential smoothing (SES, ARRSES, and HOLT’S linear). The result of data analyze is the predict error for four vaccinate data are very close to zero. SES gets 0% to 0,73%, ARRSES gets 0% to 0,66% and Holt’s linear gets 0% to 0,29%. The result of this research can conclude that exponential smoothing method effective for predict the data with trend and non-seasonal data.

Comparison Predictions of the Demam Berdarah Dengue (DBD) using Model Exponential Smoothing: Pegel’s Classification and ChatGPT

“…In Exponential smoothing method have a good model to choose the data with caracteristic trend, linearity, season, non season, and exponential. That is reinforced by previous research scientific that is using the exponential smoothing method to predict data covid vaccinate with trend characteristic got forecast error ranging from 0% to 7% [18]. Moreover, research on predictions using exponential smoothing:pegel's classification has been carried out in various areas of life, such as agricultural sector [15] that is pegel's model used is A1, with maximum error prediction is 12,216% and the minimum is 8,266%.…”