• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.349-360
• /
• 1995

This paper deals with the robustness properties of the minimum disparity estimation in linear regression models. The estimators defined as statistical quantities whcih minimize the blended weight Hellinger distance between a weighted kernel density estimator of the residuals and a smoothed model density of the residuals. It is shown that if the weights of the density estimator are appropriately chosen, the estimates of the regression parameters are robust.

• 한국지능시스템학회논문지
• /
• 제14권7호
• /
• pp.847-851
• /
• 2004

Bishop과 Nabnck에 의해 소개된 기존치 혼합 밀도 네트워크(Mixture Density Network)에서는 조건부 확률밀도 함수의 매개변수들(parameters)이 하나의 MLP(multi-layer perceptron)의 출력 벡터로 주어진다. 최근에는 변형된 혼합 밀도 네트워크(Modified Mixture Density Network)라고 하는 이름으로 조건부 확률밀도 함수의 선분포(priors), 조건부 평균(conditional means), 그리고 공분산(covariances) 등이 각각 독립적인 MLP의 출력벡터로 주어지는 경우를 다룬 연구가 보고된 바 있다. 본 논문에서는 조건부 평균이 입력에 관해 선형인 경우를 위한 버전에 대한 이론과 매트랩 프로그램 개발을 다룬다. 본 논문에서는 우선 일반적인 혼합 밀도 네트워크에 대해 간단히 설명하고, 혼합 밀도 네트워크의 출력인 다층 퍼셉트론의 매개변수를 각각 다른 다층 퍼셉트론에서 학습시키는 변형된 혼합 밀도 네트워크를 설명한 후, 각각 다른 다층 퍼셉트론을 통해 매개변수를 얻는 것은 동일하나 평균값은 선형함수를 통해 얻는 혼합 밀도 네트워크 버전을 소개한다. 그리고, 모의실험을 통하여 이러한 혼합 밀도 네트워크의 적용가능성에 대해 알아본다.

• 한국연소학회지
• /
• 제2권1호
• /
• pp.9-16
• /
• 1997

This paper investigates the linear stability of wakes with special emphasis on the difference of velocity and density. Velocity and density profiles for laminar flows have gaussian profiles. Incompressible wakes have two generalized inflection points and two unstable modes-sinuous and varicose modes. Sinuous modes are more unstable than varicose modes irrespective of density variation, which shows wakes will be destabilized by sinuous modes. Large velocity difference and density difference leads to more unstable wakes due to large momentum difference.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(2)
• /
• pp.221-224
• /
• 2002

Recently, simulation of Chemical Mechanical Polis hing is becoming more important because Process parameters on the material removal rate are complicated. And pattern-depent effects are a key concern in CMP processes. In this paper, we have been studied the changes of pattern density vs. oxide thickness with Stine's simulation model. We also have estimated the effective density using optimal window size with density mask, and have made a study of the change of oxide thickness as a function of polishing time.

• Journal of applied mathematics & informatics
• /
• 제9권1호
• /
• pp.231-238
• /
• 2002

In this paper, we study cyclic codes detecting a subclass of close-closed loop bursts viz. low-density close-closed loop bursts. A subclass of CT close-closed loop berets called CT low-density close-closed loop bursts is also studied.

• Journal of the Korean Data and Information Science Society
• /
• 제6권1호
• /
• pp.85-96
• /
• 1995

The purpose of this paper is to propose the kernel density estimator and the jackknife kernel density estimator in the presence of k's unidentified outliers, and to compare the small sample performances of the proposed estimators in a sense of mean integrated square error(MISE).

• 펄프종이기술
• /
• 제38권5호
• /
• pp.66-73
• /
• 2006

The effects of active packaging on the surface stiffness, mass, volume, density and weight changes of fresh as well as stored eggplant were studied at 11 and $25^{\circ}C$ for 10 days with active packaging material Type 1 and 2 and control. Mass, volume, and surface stiffness of eggplant decreased linearly throughout the storage period regardless of storage conditions; while the mass density showed a reverse trend in the ease of $11^{\circ}C$ storage. Reduction rate of mass, mass density and weight was observed minimum at $25^{\circ}C$ storage temperature with active packaging Type 1. The weight of eggplant decreased at a higher rate in the initial 4 days compared to that in the later period of storage regardless of storage temperature and type of packaging.

• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.483-497
• /
• 2001

When one fits a parametric density function to a data set, it is usually advisable to test the goodness of the postulated model. In this paper we study the nonparametric tests for testing the null hypothesis against general alternatives, when the null hypothesis specifies the density function up to unknown parameters. We modify the test statistic which was proposed by the first author and his colleagues. Asymptotic distribution of the modified statistic is derived and its performance is compared with some other tests through simulation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권1호
• /
• pp.126-148
• /
• 2015

Different from the existing works on coverage problems in wireless sensor networks (WSNs), this paper considers the exposure-path prevention problem by using the percolation theory in three dimensional (3D) WSNs, which can be implemented in intruder detecting applications. In this paper, to avoid the loose bounds of critical density, a bond percolation-based scheme is proposed to put the exposure-path problem into a 3D uniform lattice. Within this scheme, the tighter bonds of critical density for omnidirectional and directional sensor networks under random sensor deployment-a 3D Poisson process are derived. Extensive simulation results show that our scheme generates tighter bounds of critical density with no exposure path in 3D WSNs.

• 전기학회논문지
• /
• 제63권6호
• /
• pp.746-751
• /
• 2014

This paper presents a method to calculate a sag cumulative table and density table based on the area of vulnerability to voltage sags. The expected sag performance at sensitive load points can be easily assessed through the calculation of the sag cumulative table and density table. However, the calculation of the tables regarding large power systems is very difficult task. Therefore, an efficient method for calculating the tables is needed. In this paper, the improved method to determine the areas of vulnerability is addressed and the method to calculate the tables based on the area of vulnerability is also described. The proposed scheme can be efficiently used to assess the voltage sag performance at sensitive load points in large power systems.