Holt-Winters Forecasting: Some Practical Issues

Chatfield, Chris; Yar, Mohammad Shahar

doi:10.2307/2348687

Cited by 202 publications

(147 citation statements)

References 41 publications

(67 reference statements)

Supporting

Mentioning

139

Contrasting

Unclassified

Order By: Relevance

“…As this method is suited to seasonal series, the data was not deseasonalised beforehand. Chatfield and Yar (1988) Makridakis et al, 1982). Another approach is to use a simple linear regression on time to produce initial slope and trend (Gardner and McKenzie, 1985).…”

Section: Description Of the Studymentioning

confidence: 99%

Exponential smoothing with a damped multiplicative trend

Taylor

2003

International Journal of Forecasting

234

114

View full text Add to dashboard Cite

Multiplicative trend exponential smoothing has received very little attention in the literature.It involves modelling the local slope by smoothing successive ratios of the local level, and this leads to a forecast function that is the product of level and growth rate. By contrast, the popular Holt method uses an additive trend formulation. It has been argued that more real series have multiplicative trends than additive. However, even if this is true, it seems likely that the more conservative forecast function of the Holt method will be more robust when applied in an automated way to a large batch of series with different types of trend. In view of the improvements in accuracy seen in dampening the Holt method, in this paper, we investigate a new damped multiplicative trend approach. An empirical study, using the monthly time series from the M3-Competition, gave encouraging results for the new approach at a range of forecast horizons, when compared to the established exponential smoothing methods.

show abstract

Section: Description Of the Studymentioning

confidence: 99%

Exponential smoothing with a damped multiplicative trend

Taylor

2003

International Journal of Forecasting

234

114

View full text Add to dashboard Cite

show abstract

“…Perform the white noise test on the original sequence; The results shows the , indicate that the original sequence is a non-white noise sequence [13], which has the significance of the research. The vehicles volume shows the significant seasonal characteristics by per hour in Kunming.…”

Section: Arima Experimental Resultsmentioning

confidence: 91%

Short-term Forecast Model of Vehicles Volume Based on ARIMA Seasonal Model and Holt-Winters

Wang¹,

et al. 2017

ITM Web Conf.

View full text Add to dashboard Cite

Abstract:In order to alleviate the urban traffic congestion and ensure traffic safety, we need to do a good job in urban road traffic safety planning, make the real-time analysis and forecast of urban traffic flow to detect changes of current traffic flow in time, make scientific planning of roads and improve the road service ability and the transport efficiency of freight vehicles. The data of short-term vehicles volume is characterized by uncertainty and timing correlation series. Given this, the ARIMA seasonal model and the Holt-Winters model are used to establish a forecasting model for the short-term vehicles volume of the city. Finally, we compare the model with predictions.

show abstract

“…Após a análise descritiva foram ajustados os modelos de alisamento exponencial (Chatfield & Yar, 1988) do tipo HoltWinters aditivo, apresentando as constantes de alisamento e também seus valores previstos graficamente. O modelo HoltWinters aditivo é definido na Eq.…”

Section: Methodsunclassified

“…Tais modelos necessitam, basicamente, de um vetor de valores observados ao longo do tempo da variável climatológica de interesse. Entre os modelos tradicionais de previsão se destacam os modelos autorregressivos integrados de médias móveis (ARIMA) (Box et al, 2008) e os algoritmos de alisamento exponencial (Chatfield & Yar, 1988). Neste trabalho essas duas classes de modelos são comparadas por meio da modelagem e previsão da temperatura do ar mensal média.…”

Section: Introductionunclassified

Modelos univariados de séries temporais para previsão das temperaturas médias mensais de Erechim, RS

Chechi¹,

Bayer

2012

Rev. bras. eng. agríc. ambient.

View full text Add to dashboard Cite

R ESU M OEste trabalho apresenta uma análise de séries temporais dos dados de temperatura mínima e temperatura máxima mensal da cidade de Erechim, RS; apresenta-se uma comparação de duas classes de modelos tradicionais de previsão, nomeadamente: modelos da classe ARIMA e modelos de alisamento exponencial. Na classe de modelos ARIMA foram selecionados, utilizando-se critérios de informação, modelos do tipo SARIMA, que consideram a característica sazonal da temperatura do ar; já para os modelos de alisamento exponencial utilizaram-se os modelos Holt-Winters aditivo, em que as constantes de alisamento são determinadas de forma a minimizar o erro quadrático médio entre valores previstos e observados; esta análise permitiu a identificação de componentes como sazonalidade e períodos atípicos. Os modelos de previsão foram comparados para diferentes horizontes de previsão, sendo que os modelos da classe ARIMA se mostraram mais acurados. Os modelos ajustados se mostraram adequados para traçar previsões das variáveis de temperatura do ar, mostrando-se importantes ferramentas para a climatologia agrícola. Palavras-chave: ARIMA, alisamento exponencial, modelos de previsão, sazonalidadeUnivariate time series methods for forecasting the monthly mean air temperature in Erechim, RS A B ST R A C T This paper presents a time series analysis of the minimum and maximum air temperature of Erechim, RS. A comparison between two traditional classes of the forecasting models, namely: ARIMA models and exponential smoothing models is also presented. In the class of ARIMA models using criteria information, SARIMA type models that consider the seasonal characteristics of air temperature were selected, whereas for exponential smoothing models Holt-Winters additive algorithm were used. Smoothing constants are determined to minimize the mean square error between observed and predicted values. This analysis allowed the identification of components such as seasonality and atypical periods. The model predictions were compared for different forecast horizons. The ARIMA class models proved to be more accurate while the adjusted models were adequate for adjusting forecasts of variables of air temperature, being important tools for agricultural climatology.

show abstract

Holt-Winters Forecasting: Some Practical Issues

Cited by 202 publications

References 41 publications

Exponential smoothing with a damped multiplicative trend

Exponential smoothing with a damped multiplicative trend

Short-term Forecast Model of Vehicles Volume Based on ARIMA Seasonal Model and Holt-Winters

Modelos univariados de séries temporais para previsão das temperaturas médias mensais de Erechim, RS

Contact Info

Product

Resources

About