• Title/Summary/Keyword: Random walk model

Search Result 130, Processing Time 0.026 seconds

DIFFERENTIAL LEARNING AND ICA

  • Park, Seungjin
    • Proceedings of the Korean Institute of Intelligent Systems Conference
    • /
    • 2003.09a
    • /
    • pp.162-165
    • /
    • 2003
  • Differential learning relies on the differentiated values of nodes, whereas the conventional learning depends on the values themselves of nodes. In this paper, I elucidate the differential learning in the framework maximum likelihood learning of linear generative model with latent variables obeying random walk. I apply the idea of differential learning to the problem independent component analysis(ICA), which leads to differential ICA. Algorithm derivation using the natural gradient and local stability analysis are provided. Usefulness of the algorithm is emphasized in the case of blind separation of temporally correlated sources and is demonstrated through a simple numerical example.

  • PDF

An overlapping decomposed filter for INS initial alignment (관성항법장치의 초기정렬을 위한 중복 분해 필터)

  • 박찬국;이장규
    • 제어로봇시스템학회:학술대회논문집
    • /
    • 1991.10a
    • /
    • pp.136-141
    • /
    • 1991
  • An Overlapping Decomposed Filter(ODF) accomplishing an initial alignment of an INS is proposed in this paper. The proposed filter improves the observable condition and reduces the filtering computation time. Its good performance has been verified by simulation. Completely observable and controllable conditions of INS error model derived from psi-angle approach are introduced under varying sensor characteristics vary. The east components of gyro and accelerometer have to be the first order markov process and the rest of them are the characteristics of the random walk or first order markov process.

  • PDF

A Wald Test for a Unit Root Based on the Symmetric Estimator

  • Jong Hyup Lee;Dong Wan SHin
    • Communications for Statistical Applications and Methods
    • /
    • v.4 no.3
    • /
    • pp.677-683
    • /
    • 1997
  • For an AR(1) model with intercept $y_t=\mu+\rho{y_{t-1}}+e_t$, a test for random walk hypothesis $H_0:(\mu, \rho)=(0, 1)$is proposed, which is based on the symmetric estimator. In the vicinity of the null, the test in shown to be more powerful than the test of Dickey and Fuller(1981) based on the ordinary least squares estimator.

  • PDF

A Simple Lagrangian PDF Model for Wall-Bounded Turbulent Flows

  • Lee, Chang-Hoon;Kim, Byung-Gu;Kim, Nam-Hyun
    • Journal of Mechanical Science and Technology
    • /
    • v.14 no.8
    • /
    • pp.900-911
    • /
    • 2000
  • A simple Lagrangian pdf model is proposed with a new numerical algorithm for application in wall-bounded turbulent flows. To investigate the performance of the Lagrangian model, we minimize model's dependence on empirical constants by selecting the simplest model for turbulent dissipation rate. The effect of viscosity is also included by adding a Brownian random walk calculate the position of a particle. For the no-slip condition at the wall and correct nearwall behavior of velocity, we develop a new boundary treatment for the particles that strike the wall. By applying the model to a fully developed turbulent channel flow at low Reynolds number, we investigate the model's performance through comparison with direct numerical simulation result.

  • PDF

A Comparison Study of Forecasting Time Series Models for the Harmful Gas Emission (유해가스 배출량에 대한 시계열 예측 모형의 비교연구)

  • Jang, Moonsoo;Heo, Yoseob;Chung, Hyunsang;Park, Soyoung
    • Journal of the Korean Society of Industry Convergence
    • /
    • v.24 no.3
    • /
    • pp.323-331
    • /
    • 2021
  • With global warming and pollution problems, accurate forecasting of the harmful gases would be an essential alarm in our life. In this paper, we forecast the emission of the five gases(SOx, NO2, NH3, H2S, CH4) using the time series model of ARIMA, the learning algorithms of Random forest, and LSTM. We find that the gas emission data depends on the short-term memory and behaves like a random walk. As a result, we compare the RMSE, MAE, and MAPE as the measure of the prediction performance under the same conditions given to three models. We find that ARIMA forecasts the gas emissions more precisely than the other two learning-based methods. Besides, the ARIMA model is more suitable for the real-time forecasts of gas emissions because it is faster for modeling than the two learning algorithms.

Numerical Simulation of Advection and Diffusion using the Local Wind Model in Pusan Coastal Area, Korea (부산 연안역에서의 국지풍모델을 이용한 이류확산 수치모의)

  • 김유근;이화운;전병일
    • Journal of Korean Society for Atmospheric Environment
    • /
    • v.12 no.1
    • /
    • pp.29-41
    • /
    • 1996
  • The two-stage numerical model was used to study the relation between three-dimensional local wind model, advection/diffusion model of random walk method and second moment method on Pusan coastal area. The first stage is three dimensional time-dependent local wind model which gives the wind field and vertical dirrusion coefficient. The second stage is advection/diffusion model which uses the results of the first stage as input data. First, wind fields on Pusan coastal area for none synoptic scale wind showed typical land and sea breeze circulation, and convergence zone occured at 1200LST in northern of domain, in succession, moved northward of domain. Emissions from Sinpyeong industrial district were trasnported toward the inland by sea breeze during daytime, and reached the end part of domain about 1800LST. During nighttime, emissions return to sea by land breeze and vertical diffusion also contributes to upward transport. In order to use this model for forecast of air pollution concentration on the Pusan coastal area, it is necessary that computed value must be compared with measured value and wind fields model must also be dealt in detail.

  • PDF

Numerical Simulation of Dispersion of a Vast Point Source in Coastal Area using the Local Wind Model (국지풍모델을 이용한 연안지역 거대 점오염원의 이류확산 수치모의)

  • 전병일
    • Journal of Environmental Science International
    • /
    • v.7 no.4
    • /
    • pp.511-522
    • /
    • 1998
  • The two-stage numerical model was used to study the relation between three-dimensional local wind seal area for Korean peninsula. The first stave is three dimensional time-dependent local wind model which elves the wind field and vertical diffusion coefncient. The second stage is advection/duusion model which uses the results of the first stage as input data. First, wand fields on Korean peninsula for none synoptic scale wand showed typical land and sea breeze circulation, and the emitted particles were transported by sea breeze for daytime, emissions return to sea by land breeze for nighttime.

  • PDF

A Forecast of Shipping Business during the Year of 2013 (해운경기의 예측: 2013년)

  • Mo, Soo-Won
    • Journal of Korea Port Economic Association
    • /
    • v.29 no.1
    • /
    • pp.67-76
    • /
    • 2013
  • It has been more than four years since the outbreak of global financial crisis. However, the world economy continues to be challenged with new crisis such as the European debt crisis and the fiscal cliff issue of the U.S. The global economic environment remains fragile and prone to further disappointment, although the balance of risks is now less skewed to the downside than it has been in recent years. It's no wonder that maritime business will be bearish since the global business affects the maritime business directly as well as indirectly. This paper, hence, aims to predict the Baltic Dry Index representing the shipping business using the ARIMA-type models and Hodrick-Prescott filtering technique. The monthly data cover the period January 2000 through January 2013. The out-of-sample forecasting performance is measured by three summary statistics: root mean squared percent error, mean absolute percent error and mean percent error. These forecasting performances are also compared with those of the random walk model. This study shows that the ARIMA models including Intervention-ARIMA have lower rmse than random walk model. This means that it's appropriate to forecast BDI using the ARIMA models. This paper predicts that the shipping market will be more bearish in 2013 than the year 2012. These pessimistic ex-ante forecasts are supported by the Hodrick-Prescott filtering technique.

Fractals in the Spreading of Drifters: Observation and Simulation (표류부표 분산의 프랙탈 성질: 관측 및 시뮬레이션)

  • KANG, YONG Q.;LEE, MOONJIN
    • 한국해양학회지
    • /
    • v.29 no.4
    • /
    • pp.392-401
    • /
    • 1994
  • We examined the temporal characteristics of the oceanic eddy diffusion at 5 coastal regions of Korea by measuring the separation distances of multiple drifters released simultaneously at the same by the GPS and Decca transponder system. The observed variance of separation distance, for the time scales from minutes to hours, is proportional to t/SUP m/ with scaling exponent m between 1.2 and 2.0. The observed Lagrangian trajectories of drifters show fractal characteristics instead of random walk or Brown motion. As an effort toward a development of a realistic model of the oceanic eddy diffusion, we simulated the Lagrangian trajectories of drifters by fractional Brown motion (FBM) model. The observed variances of drifter separations can be generated by the FBM process provided the Hurst exponent is the same as the observed one. We further showed that the observed power law in the variance of drifter separations cannot be simulated with an ordinary Brown motion or random walk process.

  • PDF

Modeling and Performance Analysis of Zone-Based Registration Considering Cell by Cell Expansion of Location Area (셀 단위로 증가하는 위치영역을 고려한 영역기준 위치등록의 모형화 및 성능 분석)

  • 김경희;백장현;정호연
    • Journal of the Korean Operations Research and Management Science Society
    • /
    • v.28 no.3
    • /
    • pp.67-79
    • /
    • 2003
  • An efficient mobility management for mobile stations plays an important role in mobile communication networks. Even though many a mobility management schemes have been proposed, most of mobile communication networks adopt the mobility scheme based on zone-based registration. This paper studies the mobility management scheme that combines zone-based registration and 2-step selective paging. We assume cell by cell registration area, not ring by ring registration area of previous studies, and set up a new mobility model based on 2-dimensional random walk model considering the characteristic of zone-based registration to evaluate its performance exactly. We provide numerical results using proposed model to demonstrate the performance of zone-based registration and 2-step selective paging under various circumstances.