• 응용통계연구
• /
• 제34권5호
• /
• pp.849-861
• /
• 2021

본 논문은 확률적 자료연계 방법의 기본 개념과 이론적 모형을 소개하고, 실제 통계청 데이터를 사용하여 확률적 자료연계가 진행되는 과정과 원리를 보여준다. 먼저 확률적 자료연계와 결정적 자료연계와의 차이를 간단히 알아보고, 확률적 자료연계 방법론의 토대가 되는 Fellegi-Sunter 모형의 기본 구성과 관련된 모수(m-확률, u-확률), 가중치, 매치여부 판정기준에 대해 기술한다. 그리고 통계청 등록센서스와 인구총조사 자료를 이용하여 그 모형을 적용한 자료연계가 이루어지는 구체적인 과정에 대해 설명하고, 이를 통해 얻어진 연계 결과의 정확성을 살펴본다.

• Structural Engineering and Mechanics
• /
• 제17권2호
• /
• pp.267-279
• /
• 2004

This paper presents an improved Monte Carlo simulation for the probabilistic determination of initial cable forces of cable-stayed bridges under dead loads using the response surfaces method. A response surface (i.e. a quadratic response surface without cross-terms) is used to approximate structural response. The use of the response surface eliminates the need to perform a deterministic analysis in each simulation loop. In addition, use of the response surface requires fewer simulation loops than conventional Monte Carlo simulation. Thereby, the computation time is saved significantly. The statistics (e.g. mean value, standard deviation) of the structural response are calculated through conventional Monte Carlo simulation method. By using Monte Carlo simulation, it is possible to use the existing deterministic finite element code without modifying it. Probabilistic analysis of a truss demonstrates the proposed method' efficiency and accuracy; probabilistic determination of initial cable forces of a cable-stayed bridge under dead loads verifies the method's applicability.

• Communications for Statistical Applications and Methods
• /
• 제24권1호
• /
• pp.67-80
• /
• 2017

Over the last few decades, ensemble forecasts based on global climate models have become an important part of climate forecast due to the ability to reduce uncertainty in prediction. Moreover in ensemble forecast, assessing the prediction uncertainty is as important as estimating the optimal weights, and this is achieved through a probabilistic forecast which is based on the predictive distribution of future climate. The Bayesian model averaging has received much attention as a tool of probabilistic forecasting due to its simplicity and superior prediction. In this paper, we propose a new Bayesian model averaging method for probabilistic ensemble forecasting. The proposed method combines a deterministic ensemble forecast based on a multivariate regression approach with Bayesian model averaging. We demonstrate that the proposed method is better in prediction than the standard Bayesian model averaging approach by analyzing monthly average precipitations and temperatures for ten cities in Korea.

• 한국농공학회지
• /
• 제32권E호
• /
• pp.59-66
• /
• 1990

Abstract The formulation of the probabilistic finite element method was briefly reviewed. The method was implemented into a computer program for frame analysis which has the same analogy as finite element analysis. Another program for Monte Carlo simulation of finite element analysis was written. Two sample structures were assumed and analized. The characteristics of the second moment statistics obtained by the probabilistic finite element method was examined through numerical studies. The applicability and limitation of the method were also evaluated in comparison with the data generated by Monte Carlo simulation.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 춘계 학술발표회 논문집
• /
• pp.151-156
• /
• 2003

Variable selection algorithm for principal component analysis using penalized likelihood method is proposed. We will adopt a probabilistic principal component idea to utilize likelihood function for the problem and use HARD penalty function to force coefficients of any irrelevant variables for each component to zero. Consistency and sparsity of coefficient estimates will be provided with results of small simulated and illustrative real examples.

• Communications for Statistical Applications and Methods
• /
• 제11권1호
• /
• pp.49-58
• /
• 2004

In this talk we proposed an asymptotic test for dimensionality in the latent variable model for probabilistic principal component analysis with missing values at random. Proposed algorithm is a sequential likelihood ratio test for an appropriate Normal latent variable model for the principal component analysis. Modified EM-algorithm is used to find MLE for the model parameters. Results from simulations and real data sets give us promising evidences that the proposed method is useful in finding necessary number of components in the principal component analysis with missing values at random.

• Communications for Statistical Applications and Methods
• /
• 제24권2호
• /
• pp.143-154
• /
• 2017

A variable selection method based on probabilistic principal component analysis (PCA) using penalized likelihood method is proposed. The proposed method is a two-step variable reduction method. The first step is based on the probabilistic principal component idea to identify principle components. The penalty function is used to identify important variables in each component. We then build a model on the original data space instead of building on the rotated data space through latent variables (principal components) because the proposed method achieves the goal of dimension reduction through identifying important observed variables. Consequently, the proposed method is of more practical use. The proposed estimators perform as the oracle procedure and are root-n consistent with a proper choice of regularization parameters. The proposed method can be successfully applied to high-dimensional PCA problems with a relatively large portion of irrelevant variables included in the data set. It is straightforward to extend our likelihood method in handling problems with missing observations using EM algorithms. Further, it could be effectively applied in cases where some data vectors exhibit one or more missing values at random.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2010년도 춘계 학술발표회
• /
• pp.475-479
• /
• 2010

For a rational evaluation of the spatial distribution of consolidation settlement, it is necessary to adopt probabilistic method. In this study, mean and standard deviation of consolidation settlement of whole analysis region are evaluated by using the spatial distribution of consolidation layer which is estimated from kriging and statistics of soil properties. Using these results and probabilistic method, the area need to be raised the ground level for balancing the final design ground level are determined.

• 응용통계연구
• /
• 제36권4호
• /
• pp.279-294
• /
• 2023

재현자료를 생성할 때 순차회귀 다중대체(SRMI)를 이용하는 방식이 가장 널리 알려져 있으며, 이를 구현한 소프트웨어로 R-패키지 synthpop이 활용되고 있다. 본 논문에서는 확률적 주성분 분석(PPCA)을 이용하여 재현자료를 생성하는 방안을 제안하고 2개의 데이터 세트를 이용한 모의실험으로 SRMI 방식과 PPCA 방식을 비교하였다. 모의실험에서 PPCA 방식으로 생성한 재현자료는 쌍별 상관계수를 기준으로 원자료와의 유사성이 가장 우수함을 확인하였다. 향후 PPCA 방식을 이용하여 시계열 자료에 대한 재현자료 생성을 연구하고자 한다.

• 대한전자공학회논문지SP
• /
• 제37권5호
• /
• pp.94-104
• /
• 2000

차량 충돌 경보용 레이더 시스템의 개발에 있어 표적 추적의 정확도와 신뢰도는 매우 중요한 요소이다. 여러 표적을 동시에 추적할 때 중요한 것은 표적과 측정치와의 데이터 연관(data association) 이며, 부적절한 측정치가 어느 표적과 연관되면 그 표적은 트랙을 벗어나 추적능력을 잃어버릴 수 있고 심지어 다른 표적의 추적에도 영향을 줄 수 있다 지금까지 발표된 대부분의 데이터 연관 필터들은 근접하여 이동하는 표적들의 경우 이와 같은 문제점을 보여왔다 따라서, 현재 개발되고 있는 많은 알고리즘들은 이러한 데이터 연 관 문제의 해결에 초점을 맞추고 있다 본 논문에서는 순서통계(order statistics)를 이용한 새로운 다중 표적의 데이터 연관 방법에 대하여 서술하고자 한다 OSPDA와 OSJPDA로 불리는 제안된 방법은 각각 PDA 필터 또는 JPDA 필터에서 계산된 연관 확률을 이용하며 이 연관 확률을 결정 논리(dicision logic)에 의한 가중치로 함수화 하여 표적과 측정치 사이에 최적 혹은 최적 근처의(near optimal) 데이터 연관이 가능하도록 한 것이다 시뮬레이션 결과를 통해, 제안한 방법은 기존의 NN 필터, PDA 필터, 그리고 JPDA 필터의 성능과 비교 분석되었으며, 그 결과 제안한 OSPDA, OSJPDA 필터는 PDA, JPDA 필터보다 추적 정확도에 대해 각각 약 18%, 19% 이상으로 성능이 향상됨을 확인하였다 제안한 방법은 CAN을 통해 차량 엔진 등의 ECU와 통신하도록 개발된 DSP 보드를 이용하여 구현되었다