• Journal of the Korean Operations Research and Management Science Society
• /
• v.12 no.1
• /
• 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.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2003.05b
• /
• pp.759-762
• /
• 2003

• Korean Journal of Cognitive Science
• /
• v.31 no.1
• /
• pp.1-23
• /
• 2020

This paper introduces SNUDM, an analysis program for Ratcliff's diffusion model, which has been one of the most important models in cognitive psychology over the past 35 years and which has come to occupy an important place in cognitive neuroscience in recent years. The analysis tool is designed with the basic principles of easy comprehension and simplicity in use. A diffusion process was programmed as the limit of a simple random walk in a manner resembling Ratcliff & Tuerlinckx(2002). The response time distribution of the model was constructed by simulating the time taken by a random walk until it reaches a threshold with small steps. The optimal parameter values in the model are found to be the smallest value of the chi-square values obtained by comparing the resulting distribution and the experimental data using Simplex method. For simplicity and ease of use, the input file used here is created as a file containing the quantile of the reaction time, the trials and other information. The number of participants and the number of conditions required for such work programs are given in a way that answers the question. Using this analysis tool, the experimental data of Ratcliff, Gomez, & McKoon(2004) were analyzed. We found the very similar pattern of parameter values to Ratcliff et al.(2004) found. When comparing DMAT, fast-dm and SNUDM with the generated data, we found that when the number of trials is small, SNUDM estimates the boundary parameter to a value similar to fast-dm and less than the DMAT. In addition, when the number of trials was large, it was confirmed that all three tools estimate parameters similarly.

• Journal of Korean Society for Atmospheric Environment
• /
• v.18 no.2
• /
• 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.

• Journal of Soil and Groundwater Environment
• /
• v.20 no.3
• /
• 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
• /
• v.12 no.2
• /
• pp.28-36
• /
• 1987

Gaussian plume model is used to assess environmental dose for abnormal radioactive release in nuclear facility, but there has a problem to use it for complex terrain. In this report, MATTEW and WIND04 Codes which had been verified were used to calculate wind field in the complex terrain. Under the base of these codes principle, wind fields were obtained from the calculation of the finite difference approximation for advection-diffusion equations which satisfy the mass-conservative law. Particle concentrations and external doses were calculated by using PIC model which approximate the particle to radioactive cloud, and atmospheric diffusion of the particles from the random walk method. The results show that the adjusted wind fields and the distributions of the exposure dose vary with the topography of the complex terrain.

• 한국연소학회:학술대회논문집
• /
• 2004.06a
• /
• 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.

• Journal of the Korean GEO-environmental Society
• /
• v.4 no.3
• /
• pp.79-88
• /
• 2003

This paper presents a numerical study of the spatial behavior of a linear absorbing solute in a heterogeneous porous medium. The spatially correlated log-normal hydraulic conductivity field is generated in a given two-dimensional domain by using the geostatistical method (Turning Bands algorithm). The velocity vector field is calculated by applying the two-dimensional saturated groundwater flow equation to the Galerkin finite element method. The simulation of solute transport is carried out by using the random walk particle tracking model with CD(constant displacement) scheme in which the time interval is automatically adjusted. In this study, the spatial behavior of a solute is analyzed by the longitudinal center-of-mass displacement, longitudinal spatial spread moment and longitudinal plume skewness.

• Journal of Korea Planning Association
• /
• v.54 no.3
• /
• 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.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2007.05a
• /
• 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.