• 응용통계연구
• /
• 제29권1호
• /
• pp.133-144
• /
• 2016

본 논문에서는 주기적 자기회귀(periodic autoregressive) 모형에서 모수의 수를 줄이기 위한 모수 절약 주기적 자기회귀 모형을 연구하였다. 제안된 모수 절약 모형은 실증분석에서 실업률을 이용하여 기존의 계절 시계열 모형과 비교를 통하여 그 성능을 평가하였다. 모수 절약 구조를 부여하기 위하여 계절성에서 공통된 패턴을 찾아내는 방법을 사용하였으며 기존 주기적 자기회귀 모형과의 통계적 차이 유무는 LR 검정을 통해 확인하였다. 그 결과, 중장기적으로 주기적 자기회귀 모형이 기존의 계절시계열 모형보다 우수한 예측성능을 보였으며, 특히 모수 절약 주기적 자기 회귀 모형의 사용은 기존의 주기적 자기회귀 모형보다 우수한 예측성능을 나타내는 것을 확인하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제27권2호
• /
• pp.275-284
• /
• 2016

정확한 전력수요 예측은 에너지 소비를 줄이고 전력수급의 불균형을 방지한다. 본 연구는 외부요인의 영향을 가장 적게 받는 특정 시간대의 일 단위 전력 수요량을 참조선 (reference line)으로 한 시계열모형을 세우고자 한다. 고려된 시계열모형은 슬라이딩 창을 이용한 이중 계절성 Holt-Winters 모형과 TBATS 모형이다. 시계열모형의 모수는 2009년 1월 4일부터 2011년 12월 31일까지 자료를 이용하여 추정되었으며, 2012년 1월 1일부터 2012년 12월 29일까지의 각 모형의 전력수요량을 예측하여 성능을 비교하였다. RMSE와 MAPE를 통해 예측 성능을 비교한 결과 TBATS 모형의 성능이 우수하였다.

• 유통과학연구
• /
• 제14권11호
• /
• pp.83-90
• /
• 2016

Purpose - During the last two years, convenient stores (CS) are emerging as one of the most fast-growing retail trades in Korea. The goal of this work is to forecast and to analyze sales at CS using ARIMA-Intervention model (IM) and exponential smoothing method (ESM), together with sales at supermarkets in South Korea. Considering that two retail trades above are homogeneous and comparable in size and purchasing items on off-line distribution channel, individual behavior and characteristic can be detected and also relative superiority of future growth can be forecasted. In particular, the rapid growth of sales at CS is regarded as an everlasting external event, or step intervention, so that IM with season variation can be examined. At the same time, Winters ESM can be investigated as an alternative to seasonal ARIMA-IM, on the assumption that the underlying series shows exponentially decreasing weights over time. In case of sales at supermarkets, the marked intervention could not be found over the underlying periods, so that only Winters ESM is considered. Research Design, Data, and Methodology - The dataset of this research is obtained from Korean Statistical Information Service (1/2010~7/2016) and Survey of Service Trend of Korea Statistics Administration. This work is exploited time series analyses such as IM, ESM and model-fitting statistics by using TSPLOT, TSMODEL, EXSMOOTH, ARIMA and MODELFIT procedures in SPSS 23.0. Results - By applying seasonal ARIMA-Intervention model to sales at CS, the steep and persisting increase can be expected over the next one year. On the other hand, we expect the rate of sales growth of supermarkets to be lagging and tied up constantly in the next 2016 year. Conclusions - Based on 2017 one-year sales forecasts for CS and supermarkets, we can yield the useful information for the development of CS and also for all retail trades. Future study is needed to analyze sales of popular items individually such as tobacco, banana milk, soju and so on and to get segmented results. Furthermore, we can expand sales forecasts to other retail trades such as department stores, hypermarkets, non-store retailing, so that comprehensive diagnostics can be delivered in the future.

• 응용통계연구
• /
• 제31권2호
• /
• pp.189-204
• /
• 2018

본 연구는 2014년 1월부터 2017년 4월까지 광, 공업용 제조업을 하는 건물(GGM)의 전기 사용량에 대한 예측을 살펴보고자 한다. SARIMA, SARIMA + GARCH, Holt-Winters 방법, Fourier 변환으로 분해를 한 ARIMA 모형을 중심으로 네 가지 모형에 대한 적합을 하였다. 또한 2017년 5월 사용량에 대한 예측하고, 실제값을 고려하여 각 모형에 대해 예측 제곱근 평균 제곱 오차와 예측 오차율을 비교하였다. GGM 건물의 전기 사용량에 대한 변동이 심하기 때문에 여러 가지 모형 중에서도 변동성과 주기를 함께 고려한 SARIMA + GARCH 모형의 적합과 예측이 가장 뛰어난 것을 확인하였다.

• 한국항해항만학회지
• /
• 제32권6호
• /
• pp.425-431
• /
• 2008

본 연구에서는 광양항의 장래 컨테이너 물동량 및 교통량을 일변량 시계열모형을 통해 예측하고, 컨테이너 선박교통량을 산출하였다. 광양항의 물돌량과 입항 척당 물동량의 시계열 모형은 모두 추세와 계절적 변동이 있는 Winters 가법 모형으로 최적합 되었다. 광양항의 컨테이너 물동량은 2007년과 비교하여 2011년과 2015년에 각각 7.4%, 16.2% 가량 증가하여 약 2,756천TEU, 4,470천TEU가 될 것으로 예측되었다. 또한 2011년과 2015년의 컨테이너 입항 척당 평균 물동량은 2007년 대비 약 30.3%, 54.6% 증가하여 각각 675TEU, 801TEU가 될 것으로 예측되었다. 광양항에 대한 컨테이너 선박의 교통량은 2011년과 2015년에 각각 4,078척, 5,921척이 될 것으로 추정되었다.

• 대기
• /
• 제23권1호
• /
• pp.73-83
• /
• 2013

In recent several years, East Asia, Europe and North America have suffered successive cold winters and a number of historical records on the extreme weathers are replaced with new record-breaking cold events. As a possible explanation, several studies suggested that cryospheric conditions of Northern Hemisphere (NH), i.e. Arctic sea-ice and snow cover over northern part of major continents, are changing significantly and now play an active role for modulating midlatitude atmospheric circulation patterns that could bring cold winters for some regions in midlatitude. In this study, a dynamical seasonal prediction system for NH winter is newly developed using the snow depth initialization technique and statistically predicted sea-ice boundary condition. Since the snow depth shows largest variability in October, entire period of October has been utilized as a training period for the land surface initialization and model land surface during the period is continuously forced by the observed daily atmospheric conditions and snow depths. A simple persistent anomaly decaying toward an averaged sea-ice condition has been used for the statistical prediction of sea-ice boundary conditions. The constructed dynamical prediction system has been tested for winter 2012/13 starting at November 1 using 16 different initial conditions and the results are discussed. Implications and a future direction for further development are also described.

• 산업공학
• /
• 제16권1호
• /
• pp.16-26
• /
• 2003

In Korea, cooling power load, which occupies about 20% of peak load in 2000 and fluctuates depending on the popular usage of air conditioning systems, has been recently the focus of the load management. The first work of KEPCO (Korea Electric Power Corporation) to regulate cooling load as low as possible was to estimate its approximate scale and to develop the indirect methods to estimate it from the available time series data for the average hourly loads. However, KEPCO would like to have their methods improved both theoretically and practically. In this paper, we analyze their current indirect methods and detect their faults to design better indirect estimation methods. Under one of the assumptions of "no cooling load in April or May", the linear relationship between basic loads and GDP's, and the normalized seasonal factors of the Winters' multiplicative seasonal model, we provide ten indirect estimation methods in total and suggest the estimated cooling load(1988-1999) based on our various indirect methods.

• 융합경영연구
• /
• 제9권2호
• /
• pp.1-10
• /
• 2021

Purpose: In this work, we examine several time series models to predict internet and mobile transaction amount in South Korea, whereas Jeong (2020) has obtained the optimal forecasts for online shopping transaction amount by using time series models. Additionally, optimal forecasts based on the model considered can be calculated and applied to the Corona 19 situation. Research design, data, and methodology: The data are extracted from the online shopping trend survey of the National Statistical Office, and homogeneous and comparable in size based on 46 realizations sampled from January 2007 to October 2020. To achieve the goal of this work, both multiplicative ARIMA model and Holt-Winters Multiplicative seasonality method are taken into account. In addition, goodness-of-fit measures are used as crucial tools of the appropriate construction of forecasting model. Results: All of the optimal forecasts for the next 12 months for two online shopping transactions maintain a pattern in which the slope increases linearly and steadily with a fixed seasonal change that has been subjected to seasonal fluctuations. Conclusions: It can be confirmed that the mobile shopping transactions is much larger than the internet shopping transactions for the increase in trend and seasonality in the future.

• 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.

• 한국조리학회지
• /
• 제15권2호
• /
• pp.231-242
• /
• 2009

호텔 연회에서 가장 중요한 정보 중 하나는 매출액 자료이다. 매출액 예측은 비용을 절감시키고 인력 배분의 효율성을 증가시키고 급변하는 환경에서 경쟁하는 능력을 향상시키는 데 도움이 되는 정보를 제공한다. 본 연구는 국내외 연구에서 적합한 예측모형으로 평가되고 있는 ARIMA 모형을 이용하여 호텔 연회장의 매출액을 예측하였다. 분석을 위해서 사용한 자료는 서울 소재 GI 호텔 연회장의 월별 매출액 자료를 사용하였으며, 분석 결과 SARIMA(2,1,3)(0,1,1)가 최종적으로 추정되었다. 본 연구의 시사점은 국내외 연구에서 적합한 예측모형으로 평가되고 있는 ARIMA 모델을 호텔 연회장의 월별 매출액 자료에 적용하였다는 점과 호텔 연회 실무자들에게 참고자료로 사용할 수 있는 유용한 정보를 제공하였다는 점을 들 수 있다.