Holt-Winters Exponential Smoothing Research Articles

Implementasi Holt-Winters Exponential Smoothing untuk Peramalan Harga Bahan Pangan di Kabupaten Pamekasan

Food is one of the primary needs that is needed by humans throughout their life. Every day humans need to eat to meet the nutritional needs of the body so that they can do their activities properly. That is why changes in food prices are always a concern for everyone. The fluctuation of food prices can be a determinant for everyone to choose what food they will consume, according to their financial condition. This study aims to forecast food prices in the future by using data on food prices in the past. With price forecasting, it can be useful for planning expenditures such as monthly shopping planning and determining the selling price of food. Price changes always occur following the influencing factors, such as changes in weather and an increase in the demand for several foodstuffs on major holidays. The method used in this research is the Holt-Winters Exponential Smoothing Method. This method is a method of forecasting, which in addition to paying attention to trend factors, also observes season factors (seasonal). This study only uses food prices in Pamekasan Regency. The data used is data on food prices for the period 2012-2019. The next forecast is carried out in the following year, namely 2020. The results show that forecasting using the Holt-Winters Exponential Smoothing Method has a good accuracy value with an average MAPE value of 1.2% for the Multiplicative Model and 1.02% for the Additive Model. This result shows that Additive Model is better than Multiplicative Model because it has a smaller MAPE value.

Short-Range Forecasting of Coronavirus Disease 2019 (COVID-19) During Early Onset at Various Levels of Geographic Granularity Using Seven Different Methods

Background: The coronavirus disease 2019 (COVID-19) spread rapidly across the world since its appearance in Wuhan city, Hubei province, China in December 2019. We explore the effectiveness of forecasting methods during early onset for COVID-19 case counts over the upcoming seven-days at the county, health district, and state levels. Our goal is to assess which forecasting methods can aid decision makers about anticipated one-, three-, and seven-day case growth.Methods: We apply forecasting methods across VA’s counties and independent cities, its health districts, and the state using data reported by the The New York Times. We forecast one-, three-, and seven-days forward using: (1) a naïve approach; (2) Holt-Winters exponential smoothing (HW); (3) growth rate (Growth); (4) moving average (MA); (5) autoregressive (AR); (6) autoregressive moving average (ARMA); and (7) autoregressive integrated moving average (ARIMA). Using Median Absolute Error (MdAE) and Median Absolute Percentage Error (MdAPE), we compare the resulting 216698 forecasts to VA’s historical data through April 22, 2020.Findings: Single-day MA forecast with three-day lookback obtains the lowest MdAE and is statistically significantly different than 66 percent of evaluated alternatives at each geographic level. Furthermore, methods assuming stationary means of prior days’ case counts outperform methods with assumptions of weak- or non-stationarity. Finally, for an individual forecasting method, MdAPE evaluation reveals in most cases statistically significant differences across geographic levels demonstrating that there is not a one-size-fits-all solution to short-term COVID-19 forecasting.Interpretation: For decision makers applying short-term forecasts of COVID-19: (1) MA is an effective option for forecasting cumulative case counts one-, three-, and seven-days forward; (2) methods assuming stationarity of means in prior-observations are more effective than methods assuming weak- or non-stationary means; and (3) the intended level of geographic resolution should be a factor in selecting the appropriate forecasting method.Funding Statement: This work received no external funding.Declaration of Interests: All authors declare no competing interests.

COVID‑19 mathematical forecasting in the Russian Federation

A new coronavirus infection (CVI) is a challenge to the medical system of the Russian Federation and requires precise flow forecasting to take the necessary measures on time. The article provides an overview of modern mathematical tools for predicting the course of CVI in the world. The created CVI forecasting project office allowed to determine the most effective analysis tools in the Russian Federation — the ARIMA, SIRD and Holt–Winters exponential smoothing models. Implementation of these models allows for prediction of short-term morbidity, mortality and survival of patients with an accuracy of 99 % both in the Russian Federation in general and in the regions. In addition, the distribution of CVI was characterized. Particularly, Moscow and Moscow region have the maximum spread of infection, and other regions are lagging behind in the dynamics of the incidence by 1–3 weeks. The obtained models allow us to predict the course of the disease in the regions successfully and take the necessary measures in a timely manner.

Daily Tracking and Forecasting of the Global COVID-19 Pandemic Trend Using Holt–Winters Exponential Smoothing

Background: In December 2019, a new coronavirus disease 2019 (COVID-19) outbreak originated from a seafood market in Wuhan, China The virus, also known as SARS

Simulation of meteorological drought using exponential smoothing models: a study on Bankura District, West Bengal, India

Water scarcity and drought management is the burning issue in India and hence needs serious attention of researchers to develop rigorous plan and management. Areas that belong to various plateaus, e.g., Chotanagpur plateau, Deccan plateau, etc., are mostly affected by drought in India. In the past decade, Bankura District of West Bengal, which belongs to northeast part of Chotanagpur plateau, faced severe drought several times. However, the assessment of drought scenario in this area is far from conclusive statement till date. In this paper, we simulate standardized precipitation index (SPI) using double exponential (DE) and Holt–Winter exponential smoothing model (HW) for several time steps (e.g., 3 months, 6 months, 12 months, 24 months and 48 months) in the time period of 1979–2014. The comparative analysis between two models indicates that DE is more accurate one. DE is observed with relatively low root mean squared error (RMSE) and high R2 value. Furthermore, drought-prone zones are demarcated using combined scores of principal component analysis (PCA) and those combined scores are estimated using actual, HW and DE simulated SPI in several time steps. At the shorter (3 and 6 months) and longer time step (12, 24 and 48 months), the PCA demonstrates almost same results. The western and northwestern blocks of the district are severely affected by drought, and the southern portions are at mild condition. Spatially distributed RMSE in every time steps is also high in northwestern portions of the study region. Our result may be useful to understand the pattern of drought to take necessary action in management of water resources in Bankura District, West Bengal. Moreover, the study uses an unique methodology to simulate and assess meteorological drought, which is applicable in any region of the world.

PERAMALAN KEDATANGAN WISATAWAN MANCANEGARA INDONESIA: METODE HOLT’S WINTER EXPONENTIAL SMOOTHING

This reseacrh aims to predict the arrival of foreign tourists in Indonesia using the Exponential Smoothing method. This research is quantitative descriptive. The data used are data of foreign tourist arrivals according to nationality taken from the Badan Pusat Statistik. Data is managed through the Microsoft Excel application. In determining RMSE, Solver Parameter help is used in Microsoft Excel to determine the lowest error rate. The data used in this research indicate that there are trend and seasonal patterns, so the most suitable Exponential Smoothing method is the Holt's Winter Exponential Smoothing method. The results of this research indicate that foreign tourist arrivals in Indonesia are predicted to increase in 2020. The results of this research are expected to help the government and related agencies in planning and decision making in the tourism industry.

Peramalan harga beras IR64 kualitas III menggunakan metode Multi Layer Perceptron, Holt-Winters dan Auto Regressive Integrated Moving Average

This paper aims to present the analysis of price movements of IR64 quality III at the Cipinang Rice Main Market (PIBC) and the accuracy comparison of forecasting using Multi Layer Perceptron (MLP), Holt-Winters, and Auto Reggressive Integrated Moving Average (ARIMA) method. The data are daily price from 1 January 2016 to 31 May 2018 sourced from PT. Food Station. The analysis shows that the price of IR64 quality III rice tends to rise towards the end of 2016 and 2017. This is related to the decrease in the level of rice supply by January each year which encourages PT Food Station to conduct market operations to control the price of rice in the market. The results of accuracy comparison show that the MLP produces a value of Root Mean Square Error (RMSE) of 5,67, Holt-Winters exponential smoothing with trend and additive seasonal component produces a value RMSE of 70.71 and ARIMA method with parameters (1,1,2) resulted in RMSE values ​​of 58.71. The RMSE values ​​of the MLP method have smaller values ​​than the Holt Winter and ARIMA methods which indicate that the MLP method is more accurate

Holt–Winters smoothing enhanced by fruit fly optimization algorithm to forecast monthly electricity consumption

Electricity consumption forecasting is essential for intelligent power systems. In fact, accurate forecasting of monthly consumption to predict medium- and long-term demand substantially contributes to the appropriate dispatch and management of electric power systems. Most existing studies on monthly electricity consumption forecasting require large datasets for accurate prediction, which is severely undermined when scarce data are available. However, in practical scenarios, data is not always sufficient, thereby hindering the accurate forecasting of monthly electricity consumption. The Holt–Winters exponential smoothing allows to accurately forecast periodic series with relatively few training samples. Based on this method, we propose a hybrid forecasting model to predict electricity consumption. The fruit fly optimization algorithm is used to select the best smoothing parameters for the Holt–Winters exponential smoothing. We used electricity consumption data from a city in China to comprehensively evaluate the forecasting performance of the proposed model compared to similar methods. The results indicate that the proposed model can substantially improve the prediction accuracy of monthly electricity consumption even when few training samples are available. Moreover, the computation time of the proposed model is the shortest among the evaluated hybrid benchmark algorithms.

Forecasting incidence of tuberculosis cases in Brazil based on various univariate time-series models

Tuberculosis (TB) remains the world's deadliest infectious disease and is a serious public health problem. Control for this disease still presents several difficulties, requiring strategies for the execution of immediate combat and intervention actions. Given that changes through the decision-making process are guided by current information and future prognoses, it is critical that a country's public health managers rely on accurate predictions that can detect the evolving incidence phenomena. of TB. Thus, this study aims to analyze the accuracy of predictions of three univariate models based on time series of diagnosed TB cases in Brazil, from January 2001 to June 2018, in order to establish which model presents better performance. For the second half of 2018. From this, data were collected from the Department of Informatics of the Unified Health System (DATASUS), which were submitted to the methods of Simple Exponential Smoothing (SES), Holt-Winters Exponential Smoothing (HWES) and the Integrated Autoregressive Moving Average (ARIMA) model. In the performance analysis and model selection, six criteria based on precision errors were established: Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percent Error (MAPE) and Theil's U statistic (U1 and U2). According to the results obtained, the HWES (0.2, 0.1, 0.1) presented a high performance in relation to the error metrics, consisting of the best model compared to the other two methodologies compared here.

A Statistical out-of-Sample Forecast to Estimate Lithium-Ion Parameters That Determine State of Charge

Predicting state of charge (SOC) is crucial for an intelligent battery management system. Most of the existing SOC estimator models in literature use traditional methods such as bookkeeping and direct measurement, or rely on adaptive techniques such as neural networks. However, these models require the acquisition of significant experimental data on cell voltage and current, among others, to predict SOC with acceptable generalization performance. This paper proposes to compare two available statistical time-series forecasting methods, autoregressive integrated moving average (ARIMA) and Holt-Winters Exponential Smoothing (HWES), to forecast cell current and voltage of an actual 3.7V, 3.5Ah Lithium-ion battery for an unknown charging/discharging rate (C-rate) of C/10 using the values of the corresponding parameters from preceding C-rates of C/2, C/4, C/6, and C/8 obtained by performing lifecycle tests. The forecast values of current and voltage could, in practice, be used to calculate SOC using Coulomb-counting method [1]. The results are finally documented as a table in the attached figure. Proposed Work: Predicting the state of charge (SOC) of Lithium-ion batteries is critical for applications such as electric vehicles to determine the available capacity and function accordingly. Under scenarios where access to infrastructure and means to acquire experimental data on battery parameters is limited, the availability of an estimator model that is pre-trained for a given battery that can directly be used to predict SOC for any required C-rate is favorable. The paper proposes to use ARIMA and HWES algorithms to forecast current and voltage using historical time-series values of the same parameters obtained for C-rates C/2 to C/8. The voltage and current data are stationary as they both reject the null hypothesis in the Augmented Dickey-Fuller test [2] with a p-value of 0.01. Correlograms of the autocorrelation and partial autocorrelation functions for current and voltage are shown in Figure A in the attachment. The optimal values for the autoregressive integrative, and moving average terms are selected to be (34,1,4) such that their combination minimized the Akaike Information Criterion (AIC) [2]. The ACF and PACF plots for the fitted ARIMA model is illustrated in Figure B. The results of cell current and voltage predictions for C/10 made using the ARIMA model are shown in Figs. C and D, respectively, in the attachment. Results and discussion: The experimental setup comprised a 3.7V, 3.5Ah Li-ion battery that was subjected to different C-rates: C/2, C/4, C/6, C/8, and C/10, for 60 cycles each at discharging and charging currents of 850mAh and 2.6Ah for each test, respectively. The cell current and voltage values for C-rates C/2 through C/8 were augmented to form the time-series datasets to be used by ARIMA and HWES for model fitting. The corresponding datasets for C/10 were used solely during the forecasting phase, thereby making the forecasting process out-of-sample. For ARIMA, the mean squared error (MSE) between the forecast and observed values for C/10 voltage and current are 0.019 and 0.079, respectively. For HWES, the MSE values for C/10 voltage and current turned out to be higher, with 0.3996 and 1.6775, respectively. Besides MSE, three other errors, root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), are also calculated for both models in predicting C/10 current and voltage. This table is also included in the attachment as Fig. G. It can be seen that ARIMA had a superior performance than HWES for this specific problem. Conclusion and Future Work: To develop a reliable SOC estimator model for intelligent battery management systems that are capable of predicting SOC for future, unseen C-rates using the available cell current and voltage data from previous C-rates, this paper proposes to use ARIMA. Results show that between ARIMA and HWES, the two most widely used statistical time-series forecasting methods, ARIMA shows superior performance. As a future work, artificial neural network-based SOC and voltage estimation techniques will be identified.

Research on Consumption Law Prediction of aircraft spares based on Holt-Winters

In view of the trend and seasonal factors affecting the consumption of spare parts for aviation spares, the Holt-Winters exponential smoothing method was used for prediction research. The analysis results are consistent with the actual situation, and the practice proves that the method is feasible.

Trend Analysis and Spatial Prediction of Groundwater Levels Using Time Series Forecasting and a Novel Spatio-Temporal Method

Overexploitation of groundwater in the Malayer Plain has resulted in a continuous decline of groundwater levels over recent years with associated risks to water security. Effective water resource management requires the identification of the most susceptible regions and periods to such risks and, hence, spatio-temporal prediction tools of groundwater levels. For this purpose, we use 27 years of groundwater level records (between 1984 and 2012) and apply time series forecasting models including seasonal Autoregressive Integrated Moving Average (ARIMA) and Holt-Winters Exponential Smoothing (HWES). The spatial variation of groundwater levels is investigated by a novel method known as Fixed Rank Kriging (FRK). The results demonstrate that ARIMA outperforms HWES in fitting the training data. In contrast, the 95% confidence bound of ARIMA predictions is wider than that of HWES and ARIMA’s predicted seasonal cycle is weaker. The time series forecasting by a stochastic simulation indicated that if the current situation continues, the level of groundwater is expected to decline from 1635 m to about 1605 m by 2022. The FRK showed that the amount of groundwater extraction in the western part of the aquifer was higher than that of the northern and central parts.

Modelling Electricity Demand in Sudan Using Multiplicative Seasonal ARIMA and HOLT-WINTERS

Reliable forecast of energy demand represents a starting point in policy development and improvement of production and distribution facilities. This paper predicts the electricity demand in Sudan during the period from January 2006 to December 2016. For the purposes of these forecasts, multiplicative seasonal ARIMA and Holt-Winters models were used. The results show that the Holt-Winters exponential smoothing method, as opposed to one. An autoregressive model has been included in both the standard and the seasonal Holt-Winters methods to model the residuals and thus improve the forecast accuracy.

Forecasting Annual International Tourist Arrivals in Zambia Using Holt-Winters Exponential Smoothing

Tourism is one of the major contributors to foreign exchange earnings to Zambia and world economy. Annual International tourist arrivals in Zambia from 1995 to 2014 are considered in this paper. In this study we evaluated the model performance of Auto-Regressive Integrated Moving Average (ARIMA) and Holt Winters exponential smoothing (HWES). The error indicators: Mean percentage error (MPE), Mean absolute error (MAE), Mean absolute scaled error (MASE), Root-mean-square error (RMSE) and Mean absolute percentage error (MAPE) showed that HWES is an appropriate model with reasonable forecast accuracy. The HWES (α = 1, β = 0.1246865) showed smallest error than those of the ARIMA (0, 1, 2) models. Hence, the HWES (α = 1, β = 0.1246865) can be used to model annual international tourist arrivals in Zambia. Further, forecasting results give a gradual increase in annual international tourist arrivals of about 42% by 2024. Accurate forecasts are key to new investors and Policymakers. The Zambian government should use such forecasts in formulating policies and making strategies that will promote the tourism industry.

Commodity Indices Risk and Return Analysis Against Libor Benchmark

This study analyze the risk and return characteristics of commodity index investments against the LIBOR benchmark. Commodity-based asset allocation strategies can be optimized by benchmarking the risk and return characteristics of commodity indices with LIBOR index rate. In this study, we have considered agriculture, energy, and precious metals commodity indices and LIBOR index to determine the risk and return characteristics using estimation techniques in terms of expected return, standard deviation, and geometric mean. We analyzed the publicly available daily market data from 10/9/2001 to 12/30/2016 for benchmarking commodity indices against LIBOR. S&P GSCI Agriculture Index (SGK), S&P GSCI Energy Index (SGJ), and S&P GSCI Precious Metals Index (SGP) are taken to represent each category of widely traded commodities in the regression analysis. Our study uses time series data based on daily prices. Alternative forecasting methodologies for time series analysis are used to cross-check the results. The forecasting techniques used are Holt-Winters Exponential Smoothing and ARIMA. This methodology predicts forecasts using smoothening parameters. The empirical research has shown that the risk of each of the commodity index that represents agriculture, energy, and precious metals sector is smaller compared to its return, whereas LIBOR based interest rate benchmark shows higher risk compared to its return in recession, non-recession and overall periods.
 JEL Classification: C43, G13, G15

PERBANDINGAN PERAMALAN HASIL PRODUKSI IKAN MENGGUNAKAN METODE PERMULUSAN EKSPONENSIAL HOLT-WINTERS DAN ARIMA

Every year fish consumption in Indonesia always increases. Basis landing of fish (PPI) Kedonganan is one of the ports where the ships docked and take part in economic activities of fisheries in Bali. The aim of this study is to forecast the amount of fish production so that the fish production can be managed. Some methods that can handle seasonal factors in time series forecasting include Holt-Winters exponential smoothing and ARIMA. The results show that ARIMA(2,0,4)×(3,1,2)12 yields RMSE is 89,358 and MAPE is 0.81% whereas the value of fish outcomes prediction using Holt-Winters exponential smoothing method yields RMSE is 119,158 and MAPE is 1.14%.

Modelling and Forecasting Zimbabwe’s Tourist Arrivals Using Time Series Method: A Case Study of Victoria Falls Rainforest

Modelling and forecasting of tourist arrivals at one of the Seven Natural Wonders of the World, the Victoria Falls Rainforest, is critical to the tourism industry and economy of Zimbabwe. The aim of this paper is to provide quantitative techniques that will help with accurate tourist arrivals forecasting, shedding light on seasonality and other patterns of tourist arrivals. A time series plot of the monthly tourist arrivals statistics from January 2006 to December 2017 availed by the Zimbabwe Tourism Authority and Zimbabwe Parks and Wildlife Management Authority shows an upward trend in tourist arrivals with large fluctuations. To tame the variance which is increasing with time, a logarithm transformation is done on the data. A SARIMA (2, 1, 0)(2, 0, 0)12 model fits well to the data and outperformed other SARIMA models and the naïve, seasonal naïve and Holt-Winters exponential smoothing models. A two-year future out-of-sample forecast is done using this model and gives reasonable forecasts that indicate a general rise in tourist arrivals. Investors, tourism managers and the government can make use of such results in order to find effective and efficient solutions to the investment, foreign currency, accommodation, transport and infrastructure development problems and other tourist-related challenges faced by Zimbabweans.

Modeling Forest Woody Biomass Availability for Energy Use Based on Short-Term Forecasting Scenarios

Bioenergy in Mexico offers a great potential as a transition strategy for introducing new energy supply chains. However, studies which focus on wood supply chains for bioenergy generation at a national level are scarce. Hence, this paper presents a model for predicting short-term availability of woody biomass for energetic use according to two scenarios. Scenario A exhibits business as usual conditions. In scenario B, the availability of forest woody biomass is improved by an increment in the areas of sustainably managed forest. The theoretical, technical and economic potentials of forest woody biomass availability for energetic use were assessed using (a) numerical modeling, (b) Holt-Winters exponential smoothing and (c) regression analyses Sustainability constraints and challenges such as soil degradation, terrain slope and mechanization level were considered. A regional case study was carried out, focusing on three species with the highest utilization rates (Pinus, Quercus and Abies). Setting the base at the year 2013, a forecast analysis for the year 2023 was performed. Under scenario B, for year 2023 a technical potential of 60.22 PJ was calculated, meaning an achievement of the goals set by the National Forestry Council regarding hectares under sustainable utilization. Furthermore, a net future value analysis was carried out to account the economic output during the forecasted period. Where comprehensive data was not available, the developed model was especially useful for predicting potentially available woody biomass for energy use.

Review of Statistical and Analytical Degradation Models for Photovoltaic Modules and Systems as Well as Related Improvements

In this work, we investigate practical approaches of available degradation models and their usage in photovoltaic (PV) modules and systems. On the one hand, degradation prediction of models is described for the calculation of degradation at system level where the degradation mode is unknown and hence the physics cannot be included by the use of analytical models. Several statistical models are thus described and applied for the calculation of the performance loss using as case study two PV systems, installed in Bolzano/Italy. Namely, simple linear regression (SLR), classical seasonal-decomposition, seasonal- and trend-decomposition using Loess (STL), Holt–Winters exponential smoothing and autoregressive integrated moving average (ARIMA) are discussed. The performance loss results show that SLR produces results with highest uncertainties. In comparison, STL and ARIMA perform with the highest accuracy, whereby STL is favored because of its easier implementation. On the other hand, if monitoring data at PV module level are available in controlled conditions, analytical models can be applied. Several analytical models depending on different degradations modes are thus discussed. A comparison study is carried out for models proposed for corrosion. Although the results of the models in question agree in explanation of experimental observations, a big difference in degradation prediction was observed. Finally, a model proposed for potential induced degradation was applied to simulate the degradation of PV systems maximum power in three climatic zones: alpine (Zugspitze, Germany), maritime (Gran Canaria, Spain), and arid (Negev, Israel). As expected, a more severe degradation is predicted for arid climates.

Comparing practice-ready forecast models for weekly and monthly fluctuations of average daily traffic and enhancing accuracy by weighting methods

Abstract Knowing daily traffic for the current year is recognized as being essential in many fields of transport analysis and practice, and short-term forecasting models offer a set of tools to meet these needs. This paper examines and compares the accuracy of three representative parametric and non-parametric prediction models, selected by the analysis of the numerous methods proposed in the literature for their good combination of forecast accuracy and ease of calibration, using real-life data on Italian motorway stretches. Non-parametric K-NN regression model, Gaussian maximum likelihood model and double seasonality Holt–Winters exponential smoothing model confirm their goodness to predict the weekly and monthly fluctuations of average daily traffic with varying degrees of performance, while maintaining an easy use in professional practice, i.e. requiring ordinary professional skills and conventional analysis tools. Since combining several prediction models can give, on average, more accuracy than that of the individual models, the paper compares two weighting methods of easy implementation and susceptible to a direct use, namely the widely used information entropy method and the less widespread Shapley value method. Despite being less common than the information entropy method, the Shapley value method proves to be more capable in better combining single forecasts and produces improvements in the predictions for test data. With these remarks, the paper might be of interest to traffic technicians or analysts, in various and not uncommon tasks they might find in their work.