Earticle

이 연구에서는 토지지목 자료를 이용하여 서산시의 토지지목별 변화를 탐지 및 예측하고, 고해수면 복원자료 및 IPCC SRES 시나리오를 바탕으로 해수면 상승에 따른 침수취약성을 평가하였다. 토지지목 예측 방법은 이중지수 평활법을 사용하였다. 연구결과, 토지지목 자료를 이용한 시계열분석에서는 논과 인공구조물, 수역, 초지, 나대지 등이 증가하였고, 반면에 산림지, 습지, 밭 등은 감소하였었다. 이중지수 평활법을 이용한 토지지목 변화를 예측한 결과, 2010년에서 2050년까지 산림지, 초지, 습지, 논, 밭의 경우 지속적으로 감소하는 반면, 수역, 인공구조물, 나대지는 계속해서 증가하는 경향을 보였다. 침수 취약성 평가에서는 인공구조물과 논, 밭, 나대지 등은 침수 피해 면적이 증가하는 반면에 산림지, 수역, 습지 등은 침수 피해 면적이 감소하는 것으로 분석되었다. 특히 농경지와 인공구조물의 침수 위험성이 증가하는데 비해, 습지, 산림지의 침수 위험성은 감소하였다. 이러한 침수 면적의 증가 및 감소 원인은 해안지역에서 이루어지고 있는 관광지 개발과 산업단지 건설 등에 따른 산업화·도시화 등의 토지이용 변화에 따른 것이다.

In this study, we detected and predicted changes in terms of land categories of Seosan-si using land category data. We also evaluated flood vulnerability due to the rise in sea level based on the past sea level recovery data and IPCC SRES scenario. As a method of predicting land category, this study applied the double exponentially smoothing method. Results of the time series analysis using land category data revealed that paddy, artificial structure, water, grass field and bare land areas increased, but forest, marsh and dry field areas decreased. The results of predicting changes in land category using double exponentially smoothing showed that forest, grass field, marsh, paddy and dry field areas will continue to decrease from 2010 to 2050, but water, artificial structure and bare land areas are expected to increase continuously. Through flood vulnerability assessment, it was analyzed that flood-affected areas in artificial structure, paddy, dry field and bare land areas increased, but those in forest, water and marsh areas decreased. In particular, the risk of flooding in agricultural land and artificial structure increased, but that in marsh and forest areas decreased. It is estimated that the increase and decrease in inundation areas is caused by changes in land use resulting from industrialization and urbanization due to the construction of industrial complexes and development of tourist resorts that takes place in coastal areas.

The major advantage of the short-term load forecasting technique using general exponential smoothing is high accuracy and operational simplicity, but it makes large forecasting error when the load changes repidly. The paper has presented new technique to improve those shortcomings, and according to forecasted the technique proved to be valid for two years. The structure of load model is time function which consists of daily-and temperature-deviation component. The average of standard percentage erro in daily forecasting for two years was 2.02%, and this forecasting technique has improved standard erro by 0.46%. As relative coefficient for daily and seasonal forecasting is 0.95 or more, this technique proved to be valid.

본 논문은 이노베이션 상태공간모형을 근간으로 기존의 지수평활법을 포괄할 수 있는 다중 계절형 모형을 소개한다. 특히 이 모형은, 기존 모형의 한계를 극복하고 동일한 계절 내의 다양성을 표현할 수 있도록 계절 성분을 행렬로 표현하는 정교한 구조를 가지고 있다. 이런 구조를 이용하면 비슷한 패턴을 가지는 계절 성분의 모수를 그룹별로 분류할 수 있다. 따라서, 다중 계절형 모형은 모수절약 원칙을 달성할 수 있으며 모형의 해석이 용이한 장점을 가지고 있을 뿐만 아니라, 잠재적으로 임의의 개수의 계절성도 수용 가능하다. 본 연구에서는 다중 계절형 모형을 이용하여 시간 단위로 관측된 한국 전력 수요량을 분석하고 예측한다. 특히, 시간별 전력 수요량의 계절성은 1일 및 1주일의 두 가지로 고려되었고 이를 토대로 유사한 요일들은 공통 계절로 그룹화하였다. 모형의 예측 성능을 평가하기 위하여 기존 지수평활법의 예측 결과와 비교하였다. 그 결과, 다중 계절형 모형이 기존 지수평활법보다 예측력이 우수함을 확인하였다.

We introduce innovations state space exponential smoothing models (ISS-ESM) that can analyze time series with multiple seasonal patterns. Especially, in order to control complex structure existing in the multiple patterns, the model equations use a matrix consisting of seasonal updating parameters. It enables us to group the seasonal parameters according to their similarity. Because of the grouped parameters, we can accomplish the principle of parsimony. Further, the ISS-ESM can potentially accommodate any number of multiple seasonal patterns. The models are applied to predict electricity demand in Korea that is observed on hourly basis, and we compare their performance with that of the traditional exponential smoothing methods. It is observed that the ISS-ESM are superior to the traditional methods in terms of the prediction and the interpretability of seasonal patterns.

음식점을 경영하는데 있어서 가장 중요한 정보중 하나는 매출액 자료이다. 매출액 예측은 비용을 절감시키고, 자원 배분의 효율성을 증가시키고, 급변하는 환경에서 경쟁하는 능력을 향상시키는 데 도움이 되는 정보를 제공한다. 본 연구는 국내외 연구에서 적합한 예측모형으로 평가되고 있는 지수평활법과 ARIMA 모형을 이용하여 음식점의 매출액을 예측하였다. 분석을 위해서 사용한 자료는 서울 시내에 소재하고 있는 평균 객단가 15,000원 수준의 스테이크를 주로 판매하는 캐주얼 dining 식당 K의 일별 매출액 자료이다. 분석 결과 홀트-윈터스지수평활법, 승법계절조정지수법 및 ARIMA 방법이 단순지수평활법이나 홀트지수평활법보다 우수한 것으로 나타났다. 이러한 결과가 얻어진 이유는 원계열에 요일별 계절성이 존재하기 때문으로서, 단순지수평활법이나 홀트지수평활법이 계절 변동을 고려하지 않기 때문에 적합하지 않은 것으로 나타난 것이다.본 연구의 의의는 다음과 같다. 첫째, 국내외 연구에서 적합한 예측모형으로 평가되고 있는 지수평활법과 ARIMA방법을 음식점의 일별 매출액 자료에 적용하였다는 점이다. 둘째, 지수평활법의 예측력이 우수한 것으로 나타났는데, 이는 국외의 여러 연구결과(Miller et al, 1991; Cranage and Andrew, 1992; Blecher and Yeh, 2004)와 일치한다. 그러므로 음식점 경영을 하는 실무자들도 고도의 수학적통계적 지식을 필요로 하는 ARIMA 모형을 사용하지 않고, 비교적 쉬운 지수평활법을 이용하여 예측하여도 정확한 결과를 얻을 수 있다는 이론적 근거를 제시하였다는 점이다.

One of the most valuable parts of information useful for hospitality industry is gross sales data. The ability to forecast sales is strategically very important. The forecasting of sales helps to cut costs, increase efficiency in allocating the resources, and improve the ability to compete in a constantly changing environment. This study tests exponential smoothing models and ARIMA model to forecast restaurant sales. The results of the study show that Winters-exponential smoothing model, seasonal exponential smoothing model and ARIMA model work better in forecasting sales than single and Holt-exponential smoothing model. There is no much difference in the ability to forecast sales between Winters-exponential smoothing model, seasonal exponential smoothing model and ARIMA model Since exponential smoothing models are typically more useful than ARIMA model in terms of time and skill levels of the users, the results of this study will be of much significance for the use of forecasting techniques in hospitality industry.

This paper is concerned with the optimality of exponential smoothing applied to the general IMA process with different moving average and differencing orders. Numerical experiments were performed for IMA(m,n) process with various combinations of m and n, and the corresponding forecast errors were compared. Results show that the higher differencing order is more critical to the optimality of exponential smoothing, i.e., the IMA process with the higher moving average order, forecasted by exponential smoothing, has comparatively smaller forecast error. If the difference between the differencing order and the moving average order becomes larger, the accuracy of forecast by exponential smoothing declines gradually.

The purpose of this study is to find a forecasting model of local tax revenues, not only an accurate but practical method, among a number of alternative ones. We apply an exponential smoothing method to forecasting local tax revenues, considering the practical limits revealed in the previous studies. The results of empirical analyses reveal that exponential smoothing method is more accurate than the existing methods used for forecasting local tax revenues both in the share tax system of central government and the budget process of local governments in Korea. According to the study, the method has three advantages as follow: First, the method can make forecasting error smaller. Second, it is a simple and practical method. Third, because it is a scientific method, the forecasting results could be more objective.

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.

Intermittent demand is a demand with a pattern in which zero demands occur frequently and non-zero demands occur sporadically. This type of demand mainly appears in spare parts with very low demand. Croston's method, which is an initiative intermittent demand forecasting method, estimates the average demand by separately estimating the size of non-zero demands and the interval between non-zero demands. Such smoothing type of forecasting methods can be suitable for mid-term or long-term demand forecasting because those provides the same demand forecasts during the forecasting horizon. However, the smoothing type of forecasting methods aims at short-term forecasting, so the estimated average forecast is a factor to decrease accuracy. In this paper, we propose a forecasting method to improve short-term accuracy by improving Croston's method for intermittent demand forecasting. The proposed forecasting method estimates both the non-zero demand size and the zero demands' interval separately, as in Croston's method, but the forecast at a future period adjusted by binomial weight according to occurrence probability. This serves to improve the accuracy of short-term forecasts. In this paper, we first prove the unbiasedness of the proposed method as an important attribute in forecasting. The performance of the proposed method is compared with those of five existing forecasting methods via eight evaluation criteria. The simulation results show that the proposed forecasting method is superior to other methods in terms of all evaluation criteria in short-term forecasting regardless of average size and dispersion parameter of demands. However, the larger the average demand size and dispersion are, that is, the closer to continuous demand, the less the performance gap with other forecasting methods.

Short term load forecasting for electric power demand is essential for stable power system operation and efficient power market operation. High accuracy of the short term load forecasting can keep the power system more stable and save the power market operation cost. We propose an optimal coefficient selection method for exponential smoothing model in short term load forecasting on weekdays. In order to find the optimal coefficient of exponential smoothing model, load forecasting errors are minimized for actual electric load demand data of last three years. The proposed method are verified by case studies for last three years from 2009 to 2011. The results of case studies show that the average percentage errors of the proposed load forecasting method are improved comparing with errors of the previous methods.

As many systems depend on electronics, concern for fault tolerance is growing rapidly. For example, a car with its steering controlled by electronics and no mechanical linkage from steering wheel to front tires(steer-by-wire) should be fault tolerant because a failure can come without any warning and its effect is devastating. In order to make system fault tolerant, there has been a body of research mainly from aerospace field. This paper presents the structure of predictive smoothing voter that can filter out most erroneous values and noise. In addition, several numerical simulation results are given where the predictive smoothing voter outperforms well-known average and median voters.

전자상거래가 활성화됨에 따라 고객 개인의 관심에 부합하는 개인화된 정보나 상품 서비스를 제공하기 위하여 시간에 따라 분할하여 연산하는 시간 연관 규칙이 최근에 등장하고 있다. 본 논문은 일반적으로 정의된 연관 규칙에 대해 시간의 변화를 고려하기 위하여 최신 데이터에 가중치를 높여 주는 지수 평활법을 적용한 연관 규칙을 정의하고 이로 탐사하는 알고리즘을 제안한다. 시뮬레이션과 적용사례를 통하여 시간에 따라 분할하여 지수 평활법을 적용한 시간 연관 규칙이 기존의 것보다 실행시간은 다소 많지만 시간을 고려한 정확한 탐색률을 갖으므로 전자 상점 현장 응용에 효과적임을 확인하였다.

As electronic commerce progresses, the temporal association rule is developed from partitioned data sets by time to offer personalized services for customer's interest. In this paper, we proposed a temporal association rule with exponential smoothing method that is giving higher weights to recent data than past data. Through simulation and case study, we confirmed that it is more precise than existing temporal association rules but consumes running time.

DDoS(Distributed Denial of Service) 공격에 사용되는 네트워크 트래픽과 정상적인 서비스를 위한 네트워크 트래픽을 구분해 내는 것은 쉽지 않다. 정상적인 패킷을 유해 트래픽으로 판단하고 유해 트래픽의 공격자의 의도대로 서비스를 못하는 경우가 발생하므로, DDoS 공격으로부터 시스템을 보호하기 위해서는 공격 트래픽에 대한 정확한 분석과 탐지가 우선되어야 한다. IPv6 환경으로 전환될 때 발생하는 유해 트래픽에 대한 연구가 미약한 상태이므로, 본 논문에서는 IPv6 환경에서 NETWOX로 공격을 수행하고 공격 트래픽을 모니터링한 후 MIB(Management Information Base)객체를 지수 평활법을 적용하여 예측치를 구한 후 임계치를 산정하여 공격을 판별하는 방법을 제안한다.

Mistaking normal packets for harmful traffic may not offer service in conformity with the intention of attacker with harmful traffic, because it is not easy to classify network traffic for normal service and it for DDoS(Distributed Denial of Service) attack. And in the IPv6 environment these researches on harmful traffic are weak. In this dissertation, hosts in the IPv6 environment are attacked by NETWOX and their attack traffic is monitored, then the statistical information of the traffic is obtained from MIB(Management Information Base) objects used in the IPv6. By adapting the ESM(Exponential Smoothing Method) to this information, a normal traffic boundary, i.e., a threshold is determined. Input traffic over the threshold is thought of as attack traffic.

본 논문에서는 조사가격에 기반한 한국감정원(KAB)의 전세가격지수가 평활화 문제를 안고 있는지 여부를 확인하기 위해, 반복매매가격지수방법으로 실거래 전세가격지수를 작성한 후, KAB 전세가격지수와 비교해 보았다. 그리고 거래량의 많고 적음에 따라 평활화 정도가 다른지를 분석하였다. 2011년부터 2017년까지의 서울시 아파트 전세가격지수를 분석 대상으로 삼았다. 분석결과, KAB 전세가격지수는 낮은 변동성을 갖고 있으며, 1개월 정도의 시차를 두고 실거래 전세가격지수를 뒤따라가는 것으로 나타났다. 그리고 지수 변동의 상당부분은 실거래 전세가격지수의 변동에 기인하는 것으로 확인되었다. 이러한 평활화 현상이 경기상황에 따라 다르게 나타나는지를 확인한 결과, KAB 전세가격지수의 증가율은 전기의 두 지수의 증가율 격차를 30%정도 반영하여 조정되는 것으로 나타났다. 특히, 전세 거래량이 평균 이상일 때, 조정속도가 더 빠른 것으로 나타났다. 이에 따라 두 지수의 격차는 전세 거래량이 평균 이상일 때 더 빨리 좁혀 지는 것으로 확인되었다. 이는 평활화현상이 시장상황에 따라 비대칭적으로 일어나며, 조사자가 과거조사가격에 편의(bias)되어 있으면서도 실거래 양에 따라 실거래가격의 반영 정도를 달리하는 것을 볼 수 있다.

To see whether the survey-based rental price index compiled by the Korea Appraisal Board(KAB) has the smoothing problem, we compare it with the transaction-based rental price index that we created on the repeat sales price index model in this paper. We further investigate if the degree of smoothing problem differs according to transaction volumes. These indexes are for the ‘Jeonse’ rental market of condominiums in Seoul from 2011 to 2017. Analysis shows that the KAB’s survey-based rental price index has low volatility and a time lag of about 1 month compared with the transaction-based rental price index. And we find that a significant portion of variations in the KAB’s survey-based rental price index reflects changes in the transaction-based rental price index. We also find that the KAB’s survey-based rental price index adjusts about 30% of the increase rate gap of the previous period between the two indexes. In particular, it adjusts faster during the periods that the transaction volumes are above average. As a result, the gap between two indexes narrows more quickly when the transaction volumes are above average. These findings imply that the smoothing effect is asymmetric depending on the market condition and that surveyors reflect the change of actual transaction prices differently depending on the transaction volumes although they anchor at previous survey prices.

The Kookmin Bank's Housing Price Index(KB-HPI) which is an appraisal-based index is one of the most reliable indices in Korea. However it seems that the KB-HPI has appraisal smoothing problems. A study has reported that the KB-HPI has low volatility and time-lag as compared to a transaction-based index unpublished yet. So, under the assumption of appraisal smoothing in the KB-HPI, we estimate the partial adjustment factor to show how appraisers use the transaction price data available. We apply two methods; the partial adjustment model and the time-varying parameter model. The partial adjustment factor is estimated at 0.4821 by the first method and 0.3491 by the second method. It means that 48% or 35% of the transaction price is reflected in the appraisal of a sample house. Following the estimation, we infer a 'true' housing price index by using the partial adjustment factor. We find that the 'true' index has started to decrease from April, 2008, as has the housing market in Korea. However, the KB-HPI has continued the upward trend until August, 2008.

본 연구에서는 시간 연관 규칙에 지수 평활법을 적용한 상품 추천 알고리즘을 제안한다. 시간 연관 규칙은 기존의 연관 규칙에 시간 개념을 적용한 연관 규칙이다. 본 연구에서는 과거 데이터 보다 최신의 데이터에 가중치를 더 부여한 지수 평활 시간 연관 규칙을 제안한다. 제안한 알고리즘은 시간 의존적인 데이터에 적용하여 시뮬레이션을 한 결과 지수 평활법을 적용한 시간 연관 규칙이 기존의 시간 연관 규칙보다 실행시간 면에서 다소 오래 걸리지만 상품 추천 측면에서 더 효과적이다.

We proposed the product recommendation algorithm mixed the temporal association rule and the exponential smoothing method. The temporal association rule added a temporal concept in a commercial association rule In this paper. we proposed a exponential smoothing temporal association rule that is giving higher weights to recent data than past data. Through simulation and case study in temporal data sets, we confirmed that it is more Precise than existing temporal association rules but consumes running time.

We investigate the smoothing phenomenon in the Kookmin Bank apartment price index, the most widely used appraisal based apartment price index in Korea using a repeat sales index based on the transaction data published by the Ministry of Land, Transport and Maritime Affairs. In summary, we find that there is a greater degree of smoothing in the appraisal based apartment price index during the up market than the down market, that is, the appraisal based apartment price index tracks the sales price based index relatively slowly during the down market while it tracks the sales price based index relatively fast during the up market. The volatility in the sales price based index is far greater than the volatility in the appraisal based price index during the down market, but not during the up market. Only about one quarter of the new sale prices are priced in the appraisal during the down market while the new sale prices account for almost half of the appraisal based apartment price index during the up market. We also show that during the down market the appraisal based apartment price index reduces the gap from the sales price based index in subsequent periods only limitedly so that the smoothing during the down market tends to be persistent. We find that the transaction volume, which is a measure of information in sale prices, has a negative influence on the gap between the appraisal based apartment price index and the sales price based index. However, even when we control for the volume effect, the down market effect is still present suggesting that the asymmetrical smoothing between the down market and the up market is caused not only by the information effect of volume but also by other factors present in the down market. Our findings are generally consistent with the asymmetrical smoothing hypothesis where the smoothing in the appraisal based apartment price index is greater during the down market than during the up market.