Comparison of Forecasting Accuracy Rate of Exponential Smoothing Method on Admission of New Students

Abstract

Forecasting is a tool or technique used to predict or predict a value in the future by paying attention to relevant data or information, both past data or information as well as current data or information. There are several forecasting methods, one of which is the exponential smoothing method. In this study a comparison of forecasting accuracy to new student admission data in a study program at a university using the single exponential smoothing, double exponential smoothing and triple exponential smoothing methods. The best forecasting using the single exponential smoothing method is obtained when the parameter value α = 0.9 with the mean percentage error (MPE) = 0.0239, while the best forecasting using the double exponential smoothing method is obtained when the parameter value α = 0.8 and β = 1, with an MPE value of 0.1172. The best forecasting using the triple exponential smoothing method is obtained when the values of α = 0.6 and β = 0.9, with an MPE value of 0.0161. Based on the forecasting results of the three methods, it was concluded that the best forecasting obtained in this study was to use the triple exponential smoothing method with an MPE value of 0.0161.

Forecasting of International Flights Passenger at Soekarno-Hatta Airport using The Triple Exponential Smoothing Method

“…If the historical identification of the actual demand data shows seasonal fluctuations, it is necessary to adjust for seasonal effects by calculating the seasonal index. As an example to explain the effect of seasonality using seasonal index numbers (Himawan & Silitonga, 2020;Jonnius, 2017;Kramar & Alchakov, 2023;Triana, 2015).…”

“…Holt created this method in 1957 by adding a trend factor to the equation because the single exponential smoothing method cannot be used for trending data. The model, which is also known as a double exponential smoothing model, is essentially a single smoothing model with an additional weight that captures trends (Ginantra & Anandita, 2021;Himawan & Silitonga, 2020;Shastri et al, 2018;Singh et al, 2019). Therefore, Holt added a trend element to equation 9.…”

Forecasting of Milk Production in Northern Thailand Using Seasonal Autoregressive Integrated Moving Average, Error Trend Seasonality, and Hybrid Models

“…For ETS methods, the forecast was made considering a weighted average of past observation. The latest observation is given exponentially more weight than earlier observations ( 48 ). The state-space model of the ETS was defined as ETS (., ., .)…”

Forescating the Amount of Corn Production in North Sumatra Based on 2017 – 2021 Data Using The Single and Double Exponential Smoothing Method (Case Study of Central Bureau of Statistics of North Sumatra )

“…Single exponential smoothing will always follow every trend in real data [2], because they can do no more than regulate the upcoming forecast by a percentage of the last error. Past forecast errors are used to correct for the next forecast in the opposite direction to the error [3] This approach is subject to large forecasting errors because the errors are squared. MSE provides better accuracy than MAD so it is widely used in weighting optimization [4] discusses the Double Exponential Smoothing method, Weighted Moving Average, Moving Average, Time Series, and trend projections.…”