Modeling of inflation cases in South Sulawesi Province using single exponential smoothing and double exponential smoothing methods

Abstract

The inflation rate, particularly in South Sulawesi Province from year to year, is found to be very unstable, so that an effort to overcome the instability of the inflation rate is highly needed. One of the efforts that can be used is to carry out a process of forecasting the inflation rate, so that the government can predict the inflation rate properly in order to realize the sustainable economic growth. The aim of this study was to forecast Inflation Cases in South Sulawesi Province. The forecasting carried out in this study used the Exponential Smoothing method. Exponential Smoothing is a method that will take into account average (smoothing) the data of the past exponentially by repeating calculations continuously using the latest data. In this study, 2 Exponential Smoothing methods were compared, namely: Single Exponential Smoothing (SES) and Double Exponential Smoothing (DES) which were used to obtain prediction results and evaluate the results of predictions using the Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) methods. The smallest MAPE value was obtained when using the Single Exponential Smoothing (SES) method when the value ɑ = 0.1 with the MSE value of 0.5567 and MAPE value of 265.7126 and the Double Exponential Smoothing (DES) method when the value ɑ = 0.3 and with the MSE value of 4,256 and MAPE value of 574,519. Thus, the Single Exponential Smoothing (SES) method was regarded as the best method in predicting the inflation rate in South Sulawesi Province.

Forecasting the Tourism Development of Yunnan Province with ARIMA and Double ES

“…It is a popular statistical model commonly used for time series forecasting and analysis. [4] ARIMA models are suitable for data that exhibit non-stationarity, meaning the statistical properties of the data such as the mean and variance change over time. It is often represented as ARIMA (p, d, q), where p is the order of the autoregressive component, d is the degree of differencing, and q is the order of the moving average component.…”

“…is a method that will take into account average (smoothing) the data of the past exponentially by repeating calculations continuously using the latest data. [4]Based on the tourism development over past 30 years in Yunnan, this article aims to forecast the tourism development trend by ARIMA model and Double exponential smoothing model in the following years, from the perspective of total tourism revenue(abbreviated as TTR ).…”

The Comparison of Single and Double Exponential Smoothing Models in Predicting Passenger Car Registrations in Canada

“…The primary advantage of exponential smoothing lies in its ease of implementation and computational efficiency (Ahmar, Fitmayanti, & Ruliana, 2021;Ahmar, Rahman, Rusli, Arss, & Panday, 2023). Additionally, this method can accommodate various data components such as level, trend, and seasonality, making it a flexible tool for a wide range of forecasting situations.…”

Forecasting the Poverty Rates using Holt’s Exponential Smoothing

“…In other words, the higher the associated weight, the newer the observation. Thus, it can be concluded that, although it has similar characteristics to the ARIMA Box-Jenkins method, which produces a prediction model by calculating weighted linear summation [14], the exponential smoothing method has a different perspective of determining the weight, namely by establishing a weighting whose value explicitly decreases exponentially in the last observational data [15]. The exponential smoothing method is further divided into several methods, namely single exponential smoothing, double exponential smoothing, and triple exponential smoothing [16,17,12].…”