Earticle

만요슈(萬葉集) 4500여수의 노래들 중에는 역사적인 사건을 배경으로 한 노래들이 많이 보인다. 만요슈에 있어 정치적으로 여러 가지 사건들이 일어났던 7세기 중반부터 후반에는 동아시아의 질서를 움직이게 하는 큰 사건들이 있었다. 그 중의 하나가 나당(羅唐) 연합군에 의한 백제의 멸망과 고구려의 멸망이었고, 일본도 나당 연합군의 침공을 우려해 도읍을 이곳저곳으로 옮기는 등, 역사에 기록 될 많은 일들이 있었다. 또한 645년에 일어났던 다이카개신(大化改新)에 의한 정계 개편은 일본정치사에 큰 변화를 일으켰다. 그 중에 가장 중심에 섰던 인물이, 후에 텐치천황(天智天皇)이 된 나카노오에왕자(中大兄皇子)였다. 죠메이천황(舒明天皇)때부터 시작 된 만요슈의 노래는 여러 정치적 사건과 역사적 사건을 배경으로 하고 있다. 이러한 역사적인 사건은 그대로 문학에 투영되어 만요슈의 노래로 남게 되었다. 고토쿠천황이 결국 654년 59세의 나이로 죽으며, 아들이었던 아리마왕자(有間皇子)의 비극이 시작된다. 나카노오에왕자와 아리마왕자 두 사람의 비극은, 결국 672년의 진신난(壬申亂)의 오토모왕자(大友皇子)의 자살로 끝을 맺는다. 나카노오에왕자는　후에 텐치천황으로 화려하게 즉위하지만 결국 아리마왕자를 죽인 텐치천황의 결말은, 아리마왕자(有間皇子)가 죽은 것처럼 텐치천황의 아들인 오토모왕자의 죽음으로 끝을 맺는다. 이러한 정치적 사건의 결말이 있었던 곳이 오츠쿄(大津京)이다. 본 논문은, 이러한 역사적 배경을 둘러싼 사건 속에, 만요슈에 아리마왕자와 관련된 노래와 텐치천황의 죽음에 이르기까지의 관련된 노래들을 개략적이나마 살펴보려 한다. 또한 그 중심이었던 오츠쿄의 역사적 의미가 어떠한지를 살펴보려 한다.

Many of the 4,500 songs of Manyoshu show songs set in historical events. In the mid to late 7th century, when political events occurred in Manyoshu, there were big events that moved the order of East Asia. One of them was the fall of Baekje and Goguryeo by the Nadang allied forces, and Japan moved the capital from one place to another for fear of invasion by the Nadang allied forces. Also, the reformation of politics by the DaikanogaeShin, which occurred in 645, has caused a major change in the history of Japanese politics. One of the most important figures was Prince Nakanooe, who later became Emperor Tenchi. The song of Manyoshu, which began during the reign of Emperor Jomei, is set in various political and historical events. This historical event was projected into literature and remained a song by Manyoshu. Emperor Kotoku eventually dies in 654 at the age of 59, and the tragedy of his son Prince Arima begins. The tragedy of the two princes, Prince Nakanooe and Prince Arima, ends with the suicide of Prince Otomo of Jinshinnonan in 672. The end of Tenchicheonwang, which later ascended to the throne splendidly to Tenchicheon, but eventually killed Prince Arima, ended up with the death of Prince Otomo, the son of Emperor Tenchi, just as Prince Arima died. It was in Otsukyo that the political scandal ended. This paper, in the midst of the events surrounding this historical background, tries to outline the songs related to Prince Arima in Manyoshu and to the death of Emperor Tenchic. We will also look at the historical significance of the center, which was the historical significance of Otsu.

萬葉集4500数の歌の中には歷史的な事件を背景にした歌が多く見られる。政治的のいろいろな出来事が起きた7世紀半ばから後半には、東アジアの秩序を動かす大きな事件があった。その中の一つが、羅唐連合軍による百済の滅亡と高句麗の滅亡であり、日本も羅唐連合軍の侵攻を懸念して都をあちこちに移すなど、歴史的に記録される多くのことがあった。また、645年に起きた大化改新による政界再編は、日本政治史に大きな変化を起こした。その中で、最も中心に立った人物が、後に天智天皇になった中大兄皇子であった。舒明天皇の時から始めた萬葉集の歌は、いくつかの政治的事件と歴史的事件を背景にしている。これらの歴史的な出来事はそのまま文学に投影され、萬葉集に歌として残るされた。 孝德天皇が654年59歳で死にともに、息子だった有間皇子の悲劇が始まる。中大兄皇子と有間皇子二人の悲劇は、最終的には672年の壬申亂の大友皇子の自殺に端を結ぶ。中大兄皇子は後、天智天皇として華やかに即位するが、しかし、最終的に有間皇子を自殺させた天智天皇の結末は、有間皇子が死んでいるよう天智天皇の息子である大友皇子の死で終わりを結ぶ。これらの政治的事件の結末があったところが大津京である。本論文では、このような歴史的背景をめぐる事件の中に、萬葉集に有間皇子と関連の歌と天智天皇の死に至るまでの関連の歌を概略で見てみようする。また、その中心であった大津京の歴史的な意味がどうなのかを見てみようとする。

겐지모노가타리의 다양한 흥미로움 중에 시대 준거가 있다. 겐지모노가타리의 시대준거에 있어서 가장 설득력 있는 설명으로 紫明抄와 河海抄와 같은 고주석서에서 말하는 延喜天暦설이 있다. 桐壷 권의 맨 첫 줄에 ‘천황의 주변에 女御와 更衣와 같은 여인들이 많았다’라고 되 어 있는 것에 근거한 설명이다. 하지만 고주석서의 설명에서도, 시대 준거는 역사적인 사실 과 모노가타리의 내용이 완전히 일치하기는 어렵고, ‘어느 정도 비슷하여 짐작이 가는 정도 의 예’라고 거론하고 있으며, 그와 같은 예는 ‘더 많이 있을 수 있다’라고 말하고 있다. 그런 점에서 松本三枝子 씨의 聖徳太子伝暦 제시는, 겐지모노가타리의 시대 준거의 범위를 延喜天 暦으로 제한하지 않고 飛鳥시대 그 이전으로까지 넓힐 수 있는 가능성을 열어 주었다는 점에 서 주목할 만하다. 본 논문에서는 아스카시대의 아리마 왕자(有間皇子)와 宇治의 八宮를 비추어 생각하여 보 았다. 万葉集에 그를 애도하는 노래가 여럿 실려 있는 것을 봐도 알 수 있듯이, 아리마 왕자 는 정쟁 다툼의 극한 긴장감 속에서 안타깝게 생을 마감해야 했던 인물이다. 한 때는 장래가 기대되는 인물로 추앙되는 듯하였으나 허무하게 정쟁의 희생양이 되어 후 회스러운 삶을 살아야 했던 八宮가 결국은 스스로 이 세상의 삶을 마무리하고 산사로 들어가 는 것을 마지막으로 이 세상에서 사라지고 마는 모습에 아리마 왕자가 이 세상을 포기하게 되는 마지막 모습이 서려 있다. 겐지모노가타리의 시대 준거를 飛鳥로까지 거슬러 올라가 생각하게 하는 이유이다.

Genjimonogatari has many components that grab the readers’ interest not only culturally but also historically. Among them, era-estimate is one of the most interesting. In the very first line of the Genjimonogatari, it is the Kiritsubo-maki, “Izureno ontokinika, Nyogo, koi, amata saburaikerunakani...”, which makes the reader wonder about “which era, there were so many court ladies, for example nyogo or koi, was.” Shimeisyo and Kakaisyo are commentaries of Genjimonogatari written during the Kamakura and Muromachi age on the basis of several reasons. So these are very well known as well as being commented upon about the era-estimate of Genjimonogatari. Both Shimeisyo and Kakaisyo mention that the Era of Genjimnogatari is Engi-Tenryaku Era, which means during Daigo-tenno and Murakami-tenno. And that comment has been standing since then. But commenting about era-estimate is virtually impossible. Because almost all conditions of era-estimate cannot be perfect. They have been based on just a few facts. So other commentaries are always open to possibilities. Era-estimate of Prince Arima to Hachinomiya of Genjimonogatari that was commented through this thesis is fully meaningful. It analyzes the result based on the circumstances and conditions surrounding Hachinomiya, who failed in the power struggle, whether he truly wanted it or not.

본 연구의 목적은 호텔리츠의 투자수익률을 향상시키기 위한 방안제시를 위해 시계열 모형을 활용하여 투자수익률을 분석 및 예측하여 그 효율성을 검증하여 호텔산업의 발전을 위한 부동산간접투자기구와 호텔리츠제도의 활성화 가능성을 제시하고자하였다. 호텔리츠의 포트폴리오 기대수익률을 산출하기 위해 미국의 호텔리츠를 대상으로 2004년 11월부터 2014년 11월까지 섹터별 월별 수익률을 분석하였다. 구체적으로 수집된 데이터를 바탕으로 호텔 리츠의 월별 수익률을 도출한 후 시계열분석인ARIMA모형을 사용하기 위해 월별데이터를 회귀식으로 추정한뒤 SPSS 22.0에서 시계열 분석을 사용하여 2016년 12월까지 수익률을 예측하였다. 예측결과는 숙박/리조트의 평균수익률은 0%, 산업/오피스는–0.05%, 오피스는 0.02%, 조립식 주택은0.89%, 쇼핑센터는 0.11%, 헬스케어는 1.15%, 주거단지는 0.07%, 산업용 복합건물은 –0.10%, 복합건물은 –0.03%, 독립형건물은 1.07%, 아파트는 0.07%로 나타났으며 이들 11개 섹터의 평균수익률은 0.29%로 수익률이 저조함을 예측할 수 있으며 이는 미국 부동산시장 뿐만 아니라 국내 부동산시장도 2017년까지 수익률이 저조할 것으로 예상되어진다.

The purpose of this study predicts the output resulted from the investment to verify the efficiency by using the time series analysis model which is the way to increase the investment earning rate of Hotel REITs. This study also suggests revitalization possibility of the indirect real estate organizations and Hotel REITs systems for the development of Hotel industry using the time analysis. To output the fortpolio expected rate of return of Hotel REITs, its earning rate by the sectors compared and analysed. First, the data was collected the Hotel REITs sample in the U, S. from November 2004 to November 2014. Second, monthly earning rate of Hotel REITs based on the collected data was induced. The monthly data from November 2004 to November 2014 was estimated by ARIMA model with SPSS 22.0. The prediction results of the average earning rate, are shown as follows 0% for Lodging/Resorts, -0.05% for Industrial/Office, 0.02% for Office, 0.89% for Manufactured homes, 0.11% for Regional malls, 1.15% for Health card, 0.07% for Residential, -0.10% for Diversified, -0.03% for Mixed, 1.07% for Free standing, 0.07% for Apartments. In conclusion, the average earing rate was low with 0.29% through the 11 sectors. This results lead to expect the low earning rate in the U, S. as well as in the domestic till 2017 from now.

This paper estimates and forecasts the Busan’s exports and imports using the monthly data for years 2000-2014. To do this, this paper uses the several seasonal multiplicative ARIMA models. Among several ARIMA models, seasonal multiplicative ARIMA model (1,0,1)×(1,0,1)12 and (1,0,1)×(1,0,1)4 are selected as the most appropriate models for the estimation of exports and imports by AIC, SC and Hannan-Quin information criterion a, respectively. According to the forecasting values of the selected seasonal multiplicative ARIMA model (1,0,1)×(1,0,1)12 and (1,0,1)×(1,0,1)4, Busan’s exports and imports will increase steadily annually for 2013-2020, but there will be some volatile variations monthly due to the seasonality. Thus, to forecast the future Busan’s exports and imports and to develop them, we need to use seasonal multiplicative ARIMA model (1,0,1)×(1,0,1)12 and (1,0,1)×(1,0,1)4 for the estimation of exports and imports,

This paper concentrates on the issue of weather data mining. We propose a ARIMA algorithm based on Hadoop framework, and implement an effective weather data analyzing and forecasting system. We present the procedure to parallelize the ARIMA algorithm in the Hadoop environment, and construct a scalable, easy-to maintain, and effective weather forecasting system. Several experiments are conducted and results show that the proposed system is highly effective in terms of data storage, management, as well as query.

We compared empirically the forecast accuracies of the LSTM model, and the ARIMA model. ARIMA model used auto.arima function. Data used in the model is 100 days. We compared with the forecast results for 50 days. We collected the stock closing prices of the top 4 companies by market capitalization in Korea such as “Samsung Electronics”, and “LG Energy”, “SK Hynix”, “Samsung Bio”. The collection period is from June 17, 2022, to January 20, 2023. The paired t-test is used to compare the accuracy of forecasts by the two methods because conditions are same. The null hypothesis that the accuracy of the two methods for the four stock closing prices were the same were rejected at the significance level of 5%. Graphs and boxplots confirmed the results of the hypothesis tests. The accuracies of ARIMA are higher than those of LSTM for four cases. For closing stock price of Samsung Electronics, the mean difference of error between ARIMA and LSTM is -370.11, which is 0.618% of the average of the closing stock price. For closing stock price of LG Energy, the mean difference is -4143.298 which is 0.809% of the average of the closing stock price. For closing stock price of SK Hynix, the mean difference is -830.7269 which is 1.00% of the average of the closing stock price. For closing stock price of Samsung Bio, the mean difference is -4143.298 which is 0.809% of the average of the closing stock price. The auto.arima function was used to find the ARIMA model, but other methods are worth considering in future studies. And more efforts are needed to find parameters that provide an optimal model in LSTM.

In this study, in order to efficiently detect data patterns and outliers in time series data, outlier detection processing is performed for each section based on a smart contract in the data preprocessing process, and parameters for the ARIMA model are determined by generating and reflecting the significance and outlierrelated parameters of the data. It was created and applied to the modified arithmetic expression to lower the data abnormality. To evaluate the performance of this study, the normality of the data was compared and evaluated when the parameters of the general ARIMA model and the ARIMA model through this study were applied, and a performance improvement of more than 6% was confirmed.

In this study, in order to efficiently detect data patterns and outliers in time series data, outlier detection processing is performed for each section based on a smart contract in the data preprocessing process, and parameters for the ARIMA model are determined by generating and reflecting the significance and outlierrelated parameters of the data. It was created and applied to the modified arithmetic expression to lower the data abnormality. To evaluate the performance of this study, the normality of the data was compared and evaluated when the parameters of the general ARIMA model and the ARIMA model through this study were applied, and a performance improvement of more than 6% was confirmed.

RMB internationalization improved in 2013 with the “Belt and Road” initiative. The scheme first regionaliz ed t he R MB i n Belt a nd R oad nations, t hen globalized i t. Chinese c ompanies’ worldwide development and foreign direct investment have increased due to the “One Belt, One Road” policy. This paper examines how Chinese OFDI affects RMB internationalization under the “Belt and Road” program. The analysis uses ARIMA and combines theoretical and empirical methodologies. The research targets Belt and Road ASEAN nations. The empirical research found that outward foreign direct investment (OFDI) may boost RMB recognition and ASEAN-China economic and financial relations. Thus, it would promote RMB acceptance and globalization in these locations.

본 논문에서는 의약품 유통량 예측을 위해 기존의 통계 방식(ARIMA)과 머신러닝 방식(Informer)을 개발하고 비교하였다. 일별 데이터의 예측에서는 머신러닝 기반의 모델이 유리하며, 월별 예측에서는 ARIMA를 활용하고 데이 터가 증가하면서 Informer로 전환하는 것이 효과적임을 발견하였다. 예측 에러율(RMSE)은 기존 방식 대비 26.6% 낮아졌으며, 예측 정확도도 13% 개선되어 86.2%의 결과를 보였다. 본 논문을 통해 통계적 방법과 머신러닝 방법을 앙상블하여 최상의 결과를 얻을 수 있다는 장점을 발견하였다. 또한 머신러닝 기반의 AI 모델은 불규칙한 상황에서 도 딥러닝 연산을 통해 최선의 결과를 도출할 수 있으며, 상용화 이후에는 데이터양이 증가함에 따라 성능이 향상될 것으로 기대된다.

In this paper, the existing statistical method (ARIMA) and machine learning method (Informer) were developed and compared to predict the distribution volume of pharmaceuticals. It was found that a machine learning-based model is advantageous for daily data prediction, and it is effective to use ARIMA for monthly prediction and switch to Informer as the data increases. The prediction error rate (RMSE) was reduced by 26.6% compared to the previous method, and the prediction accuracy was improved by 13%, resulting in a result of 86.2%. Through this thesis, we find that there is an advantage of obtaining the best results by ensembleing statistical methods and machine learning methods. In addition, machine learning-based AI models can derive the best results through deep learning operations even in irregular situations, and after commercialization, performance is expected to improve as the amount of data increases.

시장의 급속한 변화와 개별 수요자 요구의 다양화로 인하여 전통적인 예측 방식은 기업의 요구사항을 충족시키기 어렵다. 다변화하는 생산 환경에서의 올바른 수요예측은 원활한 수율관리를 위한 중요한 요소이다. 현재 산업에서 보편 적으로 사용되는 기존의 많은 예측 모델은 조금씩 기능에 제한이 있다. 제안된 모델은 각 모델이 개별적으로 더 잘 수행 하는 부분을 고려하여 이러한 한계를 극복하도록 설계 되었다. 본 논문에서는 동적 프로세스 분석에 적합한 Grey Relational 분석을 통한 변수 추출을 하고, ARIMA 예측값을 통하여 산출되는 과거 수요 데이터의 특징을 포함하는 통계적으로 예측된 데이터를 생성한다. 이후, LSTM 모델과 결합하여 신경망모델이 가지는 특성인 유연성, 장기적인 의 존성 문제를 피하도록 구성되어진 구조를 통하여 수요예측에 영향을 주는 많은 요인들을 특징을 반영하여 수요예측을 산출할 수 있다.

Traditional demand forecasting methods are difficult to meet the needs of companies due to rapid changes in the market and the diversification of individual consumer needs. In a diversified production environment, the right demand forecast is an important factor for smooth yield management. Many of the existing predictive models commonly used in industry today are limited in function by little. The proposed model is designed to overcome these limitations, taking into account the part where each model performs better individually. In this paper, variables are extracted through Gray Relational analysis suitable for dynamic process analysis, and statistically predicted data is generated that includes characteristics of historical demand data produced through ARIMA forecasts. In combination with the LSTM model, demand forecasts can then be calculated by reflecting the many factors that affect demand forecast through an architecture that is structured to avoid the long-term dependency problems that the neural network model has.

This research predicts the 12-month out-of-sample wholesale price of onion and cabbage by using price and inflow data of the Garak agricultural wholesale market from January 2004 to March 2014, and the yearly production data from Statistics Korea. The HW smoothing, ARIMA, VAR, X-12 ARIMA, and ARDL models are chosen for the analysis and their prediction power are compared. The HW smoothing, ARIMA and VAR models predicts the out-of-sample wholesale prices to range between 650 and 800 won. In the X-12 model, the wholesale prices are predicted to exceed 1000 won. For cabbage, the HW Smoothing, ARIMA, and VAR models predict the cabbage wholesale price to range between 450 and 490. The X-12 ARIMA model predicts the highest price to amount to 558 won. The ARDL model suggests that it is in September when the highest price is 780 won.

In this paper, the problem of short-term load forecasting is divided into load classification and forecasting. Load classification is needed to obtain meaningful load data as input to train forecasting models. To this end, k-NN and K-mean algorithms are presented. K-mean and k-NN algorithms can handle seasonal load classification and daily load classification, respectively. The classified load data are used to train forecasting models, which are Artificial Neural Networks, Simple Exponential Smoothing, and ARIMA models. As a real case study, we tried to forecast the electric power load of the Republic of Korea. A comparison between the classified and non-classified load forecasts demonstrates the efficiency of the proposed method.

이 연구에서는 외국인 범죄와 관련된 선행연구들을 고찰하고 그 현황을 분석함과 동시에 미래에 대하여 예측해 보았다. 외국인 범죄에 대한 사회적 관심은 학계에서의 논의들을 자극해 왔으나 그 특성 및 미래에 대한 주장들은 일관되지 못한 모습을 보여주고 있다. 최근 관련 자료들의 축적과 다양한 분석기법들의 활용 가능성 증대 등은 외국인 범죄와 관련된 횡단적ㆍ종단적 분석 가능성을 보다 넓히고 있으며 이에 따라 외국인 범죄의 현황과 미래에 대하여 보다 엄밀하게 실증 분석할 수 있게 되었다. 이에 우리나라에서의 외국인 범죄 실태를 분석한 후 미래에 대하여 예측해 보았다. 먼저, 외국인 범죄의 주요 특성들은 다음과 같이 분석되었다. 첫째, 외국인 범죄는 폭력적인 성격이 강화되고 있는 것으로 나타났다. 이는 내국인 강․폭력 범죄자 비율, 흉기 사용 비율 등과의 비교를 통해서도 확인할 수 있었다. 둘째, 외국인 범죄 피의자들은 20-40대 남성에 집중되어 있으며, 중국, 베트남, 몽골, 태국 국적 외국인들이 체류비율에 비해 높은 피의자 비율을 나타내고 있었다. 셋째, 외국인 범죄는 4대 외국인 밀집지역(이태원․구로․대림․안산)에 집중되고 있으며 이러한 현실은 외국인 범죄가 이들의 집단 거주지를 중심으로 집단화ㆍ조직화되고 있는 것으로 해석할 수 있게 한다. 다음으로 외국인 범죄의 미래를 예측한 결과 최종 예측모형은 상수항이 없는 ARIMA(1,2,0) 모형으로 선정되었고 이에 따르면 외국인 범죄는 증감을 지속하나 추세적으로는 계속 증가할 것으로 예측되었다. 이와 같은 연구결과를 토대로 향후 외국인 범죄 관련 정책이 지향해야 할 방향에 대하여 살펴보았다.

This study examined current characteristics of foreigner crime in Korea and conducted a short-term forecast of it using ARIMA model. Continuing increase in the number of foreigners has given rise to the new problem including foreigner crime. There is evidence that foreigner crime continue to be violent, international, organized, and increasing. But, previous studies of foreigner crime suffered from some methodological shortcomings, including over-reliance on qualitative studies, the absence of complicated analyses, and the failure to consider domestic crime. After reviewing prior researches, statistical analysis was implemented with Korean National Police Agency's internal data concerning foreigner crime. The results are as follow. 1) Violent crimes committed by foreigners are increasing and the violence level of them is relatively higher than those by Koreans. 2) Foreign males in their 20s to 40s are more likely to commit crimes than other groups. And people in China, Vietnam, Mongolia, and Thailand are more likely to engage in criminal activities comparing their component proportion ratio. 3) Foreigners are gathered near Seoul city and some concentration areas show the characteristics of crime hot spot. 4) Foreigner crime will increase in the future. This study shows that ARIMA(1,2,0) model is the most appropriate one and forecasts more than 30,000 total crimes in 2017. According to this analysis, some implications and measures for the successful policing activities in the field of foreigner crime are suggested.

Wang (2006) and many studies argued that the Ocean Shipping Reform Act (1998) in the US allowed confidential shipping contracts between shippers and carriers and it would actually act to lessen the market power of the shipping conferences (or cartels) in the international container shipping industry. A conjecture following such arguments is that the price would fall, reflecting the new competition in the market. In this article, we want to empirically test the conjecture: the OSRA (1998), if effective, would move freight rates downward. To do so, we collect data and test the hypothesis especially on two US-related international shipping routes: Trans-Atlantic and Trans-Pacific routes. Two types of specifications are considered: Times series OLS and ARIMA models. Especially, the ARIMA results show that the new Act lowers the price at least on some routes. Especially the effect turns out to be very significant on the eastbound leg of the Trans-Atlantic route.

Data-driven decision-making is essential in the Fourth Industrial Revolution. In particular, accurate demand forecasting plays a critical role in optimizing supply chains and improving responsiveness in the global trade environment. This study aims to compare the forecasting performance of two models: ARIMA (Auto-Regressive Integrated Moving Average), a traditional time series method, and GRU (Gated Recurrent Unit), a deep learning-based approach. Using real-world demand data from the automobile parts industry, both models were evaluated based on MAPE (Mean Absolute Percentage Error), RMSE (Root Mean Squared Error), and RMSLE (Root Mean Squared Log Error). The results show that the GRU model consistently outperforms the ARIMA model, especially for highly volatile and irregular (lumpy) demand data. These findings suggest the practical applicability of GRU in demand forecasting for industries dealing with irregular patterns, and highlight the potential of combining traditional and deep learning methods for improved forecasting accuracy in international trade settings.

To accommodate non-stationarity and strong noise in the SPI data, the research used wavelet method for de-noising and autoregressive integrated moving average model(ARIMA) for prediction. Seven-day moving averages of closing time SPI data in four Asian stock marketswereanalyzed.Empiricalresults show that after de-noising more accurate forecasting results can be obtained in developed markets. More developed market indexes seem more significant improvement; while for less developed market indexes, the improvement of de-noising is less significant. This is in accordance with current situation of market.

This research applied the model that simulates the effects of inflow water quality, treatment flow rate and outflow water quality on drinking water treatment plant. The model is not a physical chemistry model. However it can evaluate the performance of sedimentation process as a statistical model. The model used transfer function ARIMA model for the prediction of turbidity on sedimentation reservoir outflow. By using the applied model, performance evaluation was executed according to the control factor per process. Coefficients of determination of the selected model were 0.95 in case of using optimal model for the transfer function model, and predictive results were estimated 0.99. Transfer function ARIMA model can replicate dynamic status of system where it is possible to diagnose system using the model in the perspective of expressing actual operation status.

The sales data sequence of the enterprise often has the seasonal effect, this paper uses X-12-ARIMA method to analyze the enterprise sales data, to eliminate the influence of the seasonal fluctuation, and provide support for the enterprise. Supply and demand is the survival and development of every enterprise, a company's sales in a certain extent, the company's production and business activities, and business sales data will show the development and change of economic time series, but the seasonal factors of production will be more, so the study of enterprise sales data seasonal changes, the following will be a clear understanding of seasonal fluctuations in corporate sales data.