• Industrial Engineering and Management Systems
• /
• 제7권1호
• /
• pp.44-50
• /
• 2008

Focusing on the exponential smoothing method equivalent to (1, 1) order ARMA model equation, a new method of estimating smoothing constant using exponential smoothing method is proposed. This study goes beyond the usual method of arbitrarily selecting a smoothing constant. First, an estimation of the ARMA model parameter was made and then, the smoothing constants. The empirical example shows that the theoretical solution satisfies minimum variance of forecasting error. The new method was also applied to the stock market price of electrical machinery industry (6 major companies in Japan) and forecasting was accomplished. Comparing the results of the two methods, the new method appears to be better than the ARIMA model. The result of the new method is apparently good in 4 company data and is nearly the same in 2 company data. The example provided shows that the new method is much simpler to handle than ARIMA model. Therefore, the proposed method would be better in these general cases. The effectiveness of this method should be examined in various cases.

• Industrial Engineering and Management Systems
• /
• 제8권4호
• /
• pp.257-263
• /
• 2009

Focusing on the idea that the equation of exponential smoothing method (ESM) is equivalent to (1, 1) order ARMA model equation, new method of estimation of smoothing constant in exponential smoothing method is proposed before by us which satisfies minimum variance of forecasting error. Theoretical solution was derived in a simple way. Mere application of ESM does not make good forecasting accuracy for the time series which has non-linear trend and/or trend by month. A new method to cope with this issue is required. In this paper, combining the trend removal method with this method, we aim to improve forecasting accuracy. An approach to this method is executed in the following method. Trend removal by a linear function is applied to the original shipping data of consumer goods. The combination of linear and non-linear function is also introduced in trend removal. For the comparison, monthly trend is removed after that. Theoretical solution of smoothing constant of ESM is calculated for both of the monthly trend removing data and the non monthly trend removing data. Then forecasting is executed on these data. The new method shows that it is useful especially for the time series that has stable characteristics and has rather strong seasonal trend and also the case that has non-linear trend. The effectiveness of this method should be examined in various cases.

• Communications for Statistical Applications and Methods
• /
• 제12권3호
• /
• pp.743-758
• /
• 2005

This paper consider penalized likelihood regression with data from exponential family. The fast computation method applied to Gaussian data(Kim and Gu, 2004) is extended to non Gaussian data through asymptotically efficient low dimensional approximations and corresponding algorithm is proposed. Also smoothing parameter selection is explored for various exponential families, which extends the existing cross validation method of Xiang and Wahba evaluated only with Bernoulli data.

• 전자공학회논문지
• /
• 제52권11호
• /
• pp.55-65
• /
• 2015

동영상 대조비 개선 과정에서 단일 영상을 위해 연구된 대조비 개선 방법들을 사용할 수 있지만, 동영상의 연속성이 고려되지 않으면 원본 동영상에 없는 깜박임을 야기할 수 있다. 또한 동영상의 연속성을 고려하는 경우, 깜박임은 억제할 수 있지만 연속성 때문에 조명의 급격한 변화할 때 불필요한 페이드인/아웃(fade-in/out) 현상이 발생하는 단점이 발생할 수 있다. 본 논문에서는 깜박임과 페이드인/아웃 현상 없이 동영상의 대조비를 개선하는 방법을 제안한다. 제안하는 방법은 Fast Gray-Level Grouping(FGLG)를 사용하여 각 프레임의 대조비를 개선하고, 깜박임을 억제하기 위해 Exponential smoothing 필터를 사용한다. 불필요한 페이드인/아웃 현상을 억제하기 위해서는 S형 함수로 Exponential smoothing 필터의 평활화 비율을 프레임 별로 적응적으로 계산하여 적용한다. 실험에서 제안하는 방법과 기존의 방법들은 6가지 측정 기준을 적용하여 성능을 비교 및 분석한다. 실험 결과, 제안하는 방법은 영상 형태 보존을 측정하는 MSSIM과 깜박임을 측정하는 Flickering score에서 정량적으로 가장 높은 결과를 보여주었으며, 시각적인 품질 비교를 통해 조명 변화에 따른 적응적인 개선을 정성적 결과로 입증하였다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2006년도 추계학술대회
• /
• pp.165-168
• /
• 2006

Exponential smoothing methods do not adapt well to unexpected changes in underlying process. Over the past few decades a number of adaptive smoothing models have been proposed which allow for the continuous adjustment of the smoothing constant value in order to provide a much earlier detection of unexpected changes. However, most of previous studies presented ad hoc procedure of adaptive forecasting without any theoretical background. In this paper, we propose a detection-adaptation procedure applied to simple and Holt's linear method. We derive level and slope change detection statistics based on Bayesian statistical theory and present distribution of the statistics by simulation method. The proposed procedure is compared with previous adaptive forecasting models using simulated data and economic time series data.

• 대한토목학회논문집
• /
• 제33권5호
• /
• pp.2009-2019
• /
• 2013

교통량은 교통망 계획, 도로 설계, 도로 관리 등에 직접적으로 활용되는 중요한 기초자료이다. 교통량은 고정식 교통량조사 장비를 설치하여 연속적인 자료를 수집하는 상시조사와 특정일을 조사하는 수시조사로 구분되어 조사되고 있다. 상시조사의 경우 조사 지점에 설치되어 있는 장비의 고장이나 오작동 등으로 인하여 교통량 자료의 결측이 발생하며, 이러한 결측을 보정하기 위하여 다양한 방법이 적용되어 왔다. 본 연구에서는 결측 발생일 전 후의 자료를 활용하는 응용 지수평활화법을 제안하였으며, 평가 결과 교통량 변동계수가 낮은 경우 보정의 정확성이 제고됨을 알 수 있었다. 게다가 지점의 교통량 변동성이 결측 보정의 정확성에 영향을 미치는 중요한 요인으로 작용한다는 것을 확인하였다. 따라서 교통량 결측 보정의 신뢰성을 높이기 위해서는 지점별-시기별 결측 보정 방법이 달리 적용되어야 할 것이다.

• International journal of advanced smart convergence
• /
• 제4권2호
• /
• pp.46-53
• /
• 2015

This paper suggests a method of real time confidence interval estimation to detect abnormal states of sensor data. For real time confidence interval estimation, the mean square errors of the exponential smoothing method and moving average method, two of the time series analysis method, were compared, and the moving average method with less errors was applied. When the sensor data passes the bounds of the confidence interval estimation, the administrator is notified through alarms. As the suggested method is for real time anomaly detection in a ship, an Android terminal was adopted for better communication between the wireless sensor network and users. For safe navigation, an administrator can make decisions promptly and accurately upon emergency situation in a ship by referring to the anomaly detection information through real time confidence interval estimation.

• 환경영향평가
• /
• 제20권4호
• /
• pp.523-538
• /
• 2011

This study analyzes the changes in the surface area of each forest cover, based on temperature data analysis and satellite imagery as the basic methods for the impact assessment of climate change on regional units. Furthermore, future changes in the forest cover are predicted using the double exponential smoothing method. The results of the study have shown an overall increase in annual mean temperature in the studied region since 1990, and an especially increased rate in winter and autumn compared to other seasons. The multi-temporal analysis of the changes in the forest cover using satellite images showed a large decrease of coniferous forests, and a continual increase in deciduous forests and mixed forests. Such changes are attributed to the increase in annual mean temperature of the studied regions. The analysis of changes in the surface area of each forest cover using the statistical data displayed similar tendencies as that of the forest cover categorizing results from the satellite images. Accordingly, rapid changes in forest cover following the increase of temperature in the studied regions could be expected. The results of the study of the forest cover surface using the double exponential smoothing method predict a continual decrease in coniferous forests until 2050. On the contrary, deciduous forests and mixed forests are predicted to show continually increasing tendencies. Deciduous forests have been predicted to increase the most in the future. With these results, the data on forest cover can be usefully applied as the main index for climate change. Further qualitative results are expected to be deduced from these data in the future, compared to the analyses of the relationship between tree species of forest and climate factors.

• Asian Journal of Business Environment
• /
• 제7권4호
• /
• pp.35-41
• /
• 2017

Purpose - This paper aims to examine several time series models to predict sales of department stores and discount store markets in South Korea, while other previous trial has performed sales of convenience stores and supermarkets. In addition, optimal predicted values on the underlying model can be got and be applied to distribution industry. Research design, data, and methodology - Two retailing types, under investigation, are homogeneous and comparable in size based on 86 realizations sampled from January 2010 to February in 2017. To accomplish the purpose of this research, both ARIMA model and exponential smoothing methods are, simultaneously, utilized. Furthermore, model-fit measures may be exploited as important tools of the optimal model-building. Results - By applying Holt-Winters' additive seasonality method to sales of two large-scale retailing types, persisting increasing trend and fluctuation around the constant level with seasonal pattern, respectively, will be predicted from May in 2017 to February in 2018. Conclusions - Considering 2017-2018 forecasts for sales of two large-scale retailing types, it is important to predict future sales magnitude and to produce the useful information for reforming financial conditions and related policies, so that the impacts of any marketing or management scheme can be compared against the do-nothing scenario.

• 응용통계연구
• /
• 제29권3호
• /
• pp.525-537
• /
• 2016

본 연구에서는 시계열 자료의 비정상성과 비선형성과 같은 복잡성을 효과적으로 포용할 수 있는 경험적모드분해법(empirical mode decomposition; EMD)을 토대로 시계열 자료의 분석 및 예측을 위한 혼합(hybrid) 모형을 연구한다. EMD에 의하여 생성되는 내재모드함수(intrinsic mode function; IMF)는 해석 및 예측의 편리성을 개선하기 위하여 누적에너지의 개념을 사용하여 그룹화하였으며, 그룹화된 IMF 및 residue의 성분들은 그 성질에 따라서 ARIMA 모형 및 지수평활법과 결합된 혼합 모형으로 예측된다. 제안된 방법은 일별 코스피 지수의 예측을 위해서 적용하였다. 다양한 형태의 혼합 모형을 사용하여 코스피 지수를 예측하였으며 전통적인 예측 방법과 비교하였다. 분석 결과, 그룹화된 성분들은 코스피 지수의 움직임을 단기적, 중기적, 장기적으로 해석하는데 편리함을 주었으며, 그룹화된 IMF 및 residue를 각각 ARIMA 모형과 지수평활법으로 조합한 혼합 모형이 우수한 예측력을 보여주었다.