• 한국경영과학회지
• /
• 제12권1호
• /
• pp.19-26
• /
• 1987

In analyzing time series, estimating the level or the current mean of the process plays an important role in understanding its structure and in being able to make forecasts. The studies the class of time series models where the level of the process is assumed to follow a random walk and the deviation from the level follow an ARMA process. The estimation and forecasting problem in a Bayesian framework and uses the Kalman filter to obtain forecasts based on estimates of level. In the analysis of time series, we usually make the assumption that the time series is generated by one model. However, in many situations the time series undergoes a structural change at one point in time. For example there may be a change in the distribution of random variables or in parameter values. Another example occurs when the level of the process changes abruptly at one period. In order to study such problems, the assumption that level follows a random walk process is relaxed to include a major level change at a particular point in time. The major level change is detected by examining the likelihood raio under a null hypothesis of no change and an alternative hypothesis of a major level change. The author proposes a method for estimation the size of the level change by adding one state variable to the state space model of the original Kalman filter. Detailed theoretical and numerical results are obtained for th first order autoregressive process wirth level changes.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2003년도 학술발표회논문집(2)
• /
• pp.759-762
• /
• 2003

• 인지과학
• /
• 제31권1호
• /
• pp.1-23
• /
• 2020

이 논문에서는 지난 40여 년 동안 인지심리학에서 가장 중요한 모형 가운데 하나이며 근래에는 인지신경과학에서도 중요한 자리를 차지하고 있는 Ratcliff의 확산(diffusion)모형을 분석하는 도구 SNUDM을 소개한다. SNUDM은 확산과정을 Ratcliff & Tuerlinckx(2002)에 소개된 방식으로 단순 무작위걷기(random walk)를 묘사했다. 구체적으로, 모형이 생성하는 반응시간 분포는 주어진 파라미터 값들에서 작은 걸음으로 무작위걷기를 하여 일정 수준에 다다를 때까지 걸린 시간들로 이루어졌고, 모형의 파라미터 추정치는 단순도형(Simplex) 방식을 이용하여 실험 자료와 생성된 분포를 비교하기 위해 계산된 카이제곱값을 최소화하는 파라미터의 값을 사용한다. 사용의 간편함을 위해, 입력 파일은 반응시간의 분위수(quantile), 시행수와 기타 정보를 담은 파일로 간단하게 했고, 프로그램 작동에 필요한 피험자 수와 조건 수 등은 질문에 답을 하는 방식으로 입력하도록 했으며, 조건에 따라 비교할 파라미터와 그렇지 않고 고정할 파라미터도 미리 지정하도록 했다. 분석도구 SNUDM이 파라미터 값을 제대로 찾아내는지를 알아보기 위해 Ratcliff, Gomez, & McKoon(2004)의 실험1 자료를 써서 검토한 결과, 그들이 보고한 실험 조건들 사이에서 보인 상대적인 표집율의 크기에서 동일한 패턴을 얻었다. 또한 SNUDM으로 생성된 자료를 DMAT과 fast-dm의 자료와 비교해 보았을 때 SNUDM은 시행수가 적을 경우에는 경계 파라미터를 fast-dm과는 비슷한 값을 추정하였고 DMAT보다는 작은 값으로 추정했으나 시행수가 많은 경우에는 세 도구 모두 비슷하게 파라미터를 추정하는 것을 확인하였다.

• 한국대기환경학회지
• /
• 제18권2호
• /
• pp.67-83
• /
• 2002

The coupled model (SMART) of dynamic meteorology model and particle dispersion model was developed. The numerical experiment on the relationship between change of land use and diffusion behavior in complex terrain was carried out using this model. It tried to investigate the change of particle diffusion behavior and local weather under the condition in which land-land breeze and sea breeze and mountain breeze intermingled. The numerical experiment results are as follows; 1) The more complicated local circulation field of the interaction of sea breeze, mountain breeze and Land -land breeze is formed. Then, the region circulation in which the urbanization is specific by location of the region is strengthened and is weakened. 2) Though in the region with dominant sea breeze, Land-land breeze does not appear directly, the progress of the sea wind to the inland is affected. 3) In the prediction of the air diffusion, emission high quality and accurate information of the emission site are important. That is to say, the dispersion predicting result which emission high quality and small error of the site perfectly vary for Land - land breeze in the effect may be brought about.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제20권3호
• /
• pp.51-64
• /
• 2015

In the present study, we develop two history matching techniques based on Markov chain Monte Carlo method where radial basis function and Gaussian distribution generated by unconditional geostatistical simulation are employed as the random walk transition kernels. The Bayesian inverse methods for aquifer characterization as the developed models can be effectively applied to the condition even when the targeted information such as hydraulic conductivity is absent and there are transient hydraulic head records due to imposed stress at observation wells. The model which uses unconditional simulation as random walk transition kernel has advantage in that spatial statistics can be directly associated with the predictions. The model using radial basis function network shares the same advantages as the model with unconditional simulation, yet the radial basis function network based the model does not require external geostatistical techniques. Also, by employing radial basis function as transition kernel, multi-scale nested structures can be rigorously addressed. In the validations of the developed models, the overall predictabilities of both models are sound by showing high correlation coefficient between the reference and the predicted. In terms of the model performance, the model with radial basis function network has higher error reduction rate and computational efficiency than with unconditional geostatistical simulation.

• Journal of Radiation Protection and Research
• /
• 제12권2호
• /
• pp.28-36
• /
• 1987

원자력 시설에서 대기중으로 방출되는 방사성 구름에 의한 환경선량계산에는 Gaussian plume model 주로 사용되고 있으나, 바람의 분포나 대기의 흐트러짐이 공간적으로 일정하지 않은 복잡 지형에의 적용에는 문제가 있다. 복잡 지형을 고려한 기류계산에는 MATTEW, WIND04 코드가 그 타당성을 인정받고 있다. 이러한 코드의 원리를 기초로 하여, 질량보존법칙을 만족하는 이류 확산 방정식을 유한차분법으로 계산하고 풍속장을 구하였다. 입자 농도와 피폭선량은 방사성 구름을 입자군으로 근사시키는 PIC model을 이용하여 계산하였으며, 입자의 대기 확산은 Random Walk법을 이용하였다. 계산 결과, 지형에 의한 풍속, 풍향의 변화를 알 수 있었으며, 피폭선량분포를 구할 수 있었다.

• 한국연소학회:학술대회논문집
• /
• 한국연소학회 2004년도 제28회 KOSCO SYMPOSIUM 논문집
• /
• pp.102-111
• /
• 2004

Partial quenching structure of turbulent diffusion flames in a turbulent mixing layer is investigated by the method of flame hole dynamics to develope a prediction model for the turbulent lift off. The present study is specifically aimed to remedy the problem of the stiff transition of the conditioned partial burning probability across the crossover condition by adopting level-set method which describes propagating or retreating flame front with specified propagation speed. In light of the level-set simulations with two model problems for the propagation speed, the stabilizing conditions for a turbulent lifted flame are suggested. The flame hole dynamics combined with level-set method yields a temporally evolving turbulent extinction process and its partial quenching characteristics is compared with the results of the previous model employing the flame-hole random walk mapping. The probability to encounter reacting' state, conditioned with scalar dissipation rate, demonstrated that the conditional probability has a rather gradual transition across the crossover scalar dissipation rate in contrast to the stiff transition of resulted from the flame-hole random walk mapping and could be attributed to the finite response of the flame edge propagation.

• 한국지반환경공학회 논문집
• /
• 제4권3호
• /
• pp.79-88
• /
• 2003

비균질 다공성 매질을 통과하는 선형 흡착 용질의 공간적 거동에 대한 특성이 수치적 기법을 통해 분석되었다. 공간적으로 서로 상관된 투수계수의 대수적 분포를 지구통계학적 기법인 TBM(Turning Bands Method)을 사용하여 2차원 공간 내에 발생시켰으며, 이를 통한 지하수 흐름의 수리수두분포와 유속벡터장은 정상상태의 포화된 2차원 지하수 흐름 방정식에 Galerkin 유한요소법을 적용하여 계산하였다. 또한, 용질이동에 대한 수치모의는 시간간격을 자동으로 보정해주는 CD(constant displacement)기법이 포함된 난보모형(Random Walk Particle Tracking Model, RWPTM)을 통해 수행되었다. 본 연구에서 용질의 공간적 거동의 특성은 종방향 질량중심 이동거리, 종방향 공간 퍼짐 모멘트 그리고 용질분포의 종방향 왜곡계수를 통해 분석되었다.

• 국토계획
• /
• 제54권3호
• /
• pp.106-118
• /
• 2019

In order to develop pedestrian navigation service that provides optimal pedestrian routes based on pedestrian satisfaction levels, it is required to develop a prediction model that can estimate a pedestrian's satisfaction level given a certain condition. Thus, the aim of the present study is to develop a pedestrian satisfaction prediction model based on three machine learning algorithms: Logistic Regression, Random Forest, and Artificial Neural Network models. The 2009, 2012, 2013, 2014, and 2015 Pedestrian Satisfaction Survey Data in Seoul, Korea are used to train and test the machine learning models. As a result, the Random Forest model shows the best prediction performance among the three (Accuracy: 0.798, Recall: 0.906, Precision: 0.842, F1 Score: 0.873, AUC: 0.795). The performance of Artificial Neural Network is the second (Accuracy: 0.773, Recall: 0.917, Precision: 0.811, F1 Score: 0.868, AUC: 0.738) and Logistic Regression model's performance follows the second (Accuracy: 0.764, Recall: 1.000, Precision: 0.764, F1 Score: 0.868, AUC: 0.575). The precision score of the Random Forest model implies that approximately 84.2% of pedestrians may be satisfied if they walk the areas, suggested by the Random Forest model.

• 한국정보기술응용학회:학술대회논문집
• /
• 한국정보기술응용학회 2007년도 춘계학술대회
• /
• pp.47-54
• /
• 2007

In this study, we consider the movement-based registration (MBR), location-based registration (LBR) and distance-based registration (DBR) schemes. Analytical models based on a 2-dimensional random walk in a hexagonal cell configuration are considered to analyze and compare the performances of these three schemes. We focus on the derivation of the registration costs of LBR and DBR using an analytical method and then show that DBR always outperforms both MBR and LBR. Numerical results are provided to demonstrate the validity of our models under various circumstances.