• Journal of the Korean Statistical Society
• /
• 제18권1호
• /
• pp.13-25
• /
• 1989

This paper suggests a classification rule with unequal covariance matrices when a patterned incomplete data are involved in the discriminant analysis. This is an extension of Geisser's (1966) result to the case of missing observations. For the calssificaiton rule, we introduce an algorithm which contains data augmentation step and Monte Carlo integration step and show that the algorithm yields a consistant estimator of true classification probability. The proposed method is compared to the complete observation vector method through a Monte Carlo study. The results show that the suggested method, in general, performs better than the complete observation vector method which ignores those vectors of observation with one or more missing values from the analysis. The results also verify the consistency of the algorithm.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 1996년도 추계학술대회발표논문집; 고려대학교, 서울; 26 Oct. 1996
• /
• pp.31-34
• /
• 1996

The extensive researches have compared the performance of neural networks(NN) with those of various statistical techniques for the classification problem. The empirical results of these comparative studies have indicated that the neural networks often outperform the traditional statistical techniques. Moreover, there are some efforts that try to combine various classification methods, especially multivariate discriminant analysis with neural networks. While these efforts improve the performance, there exists a problem violating robust assumptions of multivariate discriminant analysis that are multivariate normality of the independent variables and equality of variance-covariance matrices in each of the groups. On the contrary, cluster analysis alleviates this assumption like neural networks. We propose a new approach to classification problems by combining the cluster analysis with neural networks. The resulting predictions of the composite model are more accurate than each individual technique.

• 한국항공우주학회지
• /
• 제34권10호
• /
• pp.48-55
• /
• 2006

본 논문에서는 다중 필터 융합을 위한 기법을 제안하고 이를 INS/GPS 결합항법장치에 적용한다. IIR 형태의 EKF와 FIR 형태의 RHKF 필터를 각 필터의 잉여값(residual)을 이용한 적응형 융합 확률을 계산하여 융합함으로써 두 필터의 장점을 갖도록 한다. 이 융합 기법을 적용한 INS/GPS 결합항법장치는 기존의 단독 필터를 사용하는 것보다 강인한 특성을 갖는 항법정보를 제공한다.

• 대한수학회논문집
• /
• 제21권4호
• /
• pp.779-786
• /
• 2006

Let $\mathbb{X}_t$ be an m-dimensional linear process defined by $\mathbb{X}_t=\sum{_{j=0}^\infty}\;A_j\;\mathbb{Z}_{t-j}$, t = 1, 2, $\ldots$, where $\mathbb{Z}_t$ is a sequence of m-dimensional random vectors with mean 0 : $m\times1$ and positive definite covariance matrix $\Gamma:m{\times}m$ and $\{A_j\}$ is a sequence of coefficient matrices. In this paper we give sufficient conditions so that $\sum{_{t=1}^{[ns]}\mathbb{X}_t$ (properly normalized) converges weakly to Wiener measure if the corresponding result for $\sum{_{t=1}^{[ns]}\mathbb{Z}_t$ is true.

• Journal of Electrical Engineering and information Science
• /
• 제3권3호
• /
• pp.373-384
• /
• 1998

This paper presents a maneuvering target model with the maneuver dynamics modeled as a jump process of Poisson-type. The jump process represents the deterministic maneuver(or pilot commands) and is described by a stochastic differential equation driven by a Poisson process taking values a set of discrete states. Employing the new maneuver model along with the noisy observations described by linear difference equations, the author has developed a new linear, recursive, unbiased minimum variance filter, which is structurally simple, computationally efficient, and hence real-time implementable. Futhermore, the proposed filter does not involve a computationally burdensome technique to compute the filter gains and corresponding covariance matrices and still be able to track effectively a fast maneuvering target. The performance of the proposed filter is assessed through the numerical results generated from the Monte-Carlo simulation.

• Journal of the Korean Statistical Society
• /
• 제16권2호
• /
• pp.92-101
• /
• 1987

Our interest is to access in some way teh relative odds or probability that a multivariate observation Z belongs to one of k multivariate normal populations with unequal covariance matrices. We derived the empirical Bayes posterior odds ratio for the classification rule when population parameters are unknown. It is a generalization of the posterior odds ratio suggested by Gelsser (1964). The classification rule does not have complicated distribution theory which a large variety of techniques from the sampling viewpoint have. The proposed posterior odds ratio is compared to the Gelsser's posterior odds ratio through a Monte Carlo study. The results show that the empiricla Bayes posterior odds ratio, in general, performs better than the Gelsser's. Especially, for large dimension of Z and small training sample, the performance is prominent.

• 응용통계연구
• /
• 제6권1호
• /
• pp.95-103
• /
• 1993

두 개의 공분산행렬 $V_1과 V_2$로 구별되는 두 개의 선형모형에서 BLUE끼리 같을 필 요충분조건이 유도된다. 그리고 이 발견으로 쉽게 이해되는 여러 응용사례도 보여준다. 그동 안 여러 논문에서 언급되어 온 BLUE와 OLSE가 같을 필요충분조건도 논의된다.

• 전자공학회논문지B
• /
• 제32B권7호
• /
• pp.989-1002
• /
• 1995

In this paper, we propose a new algorithm, so called the Signal Decorrelation via Virtual Translation of Array (SDVTA) algorithm, for estimating the directions of arrival(DOA's) of narrowband coherent signals incident on a general array. An effective procedure is composed of transforming the steering matrix of the original array into that of the virtually translated sensor array and taking the average of the transformed covariance matrices in order to decorrelate the coherent signals. The advantage of this approach is in that 1) it can estimate the DOA's of m-1 coherent signals(M : the number of array sensors) since the effective aperture size is never reduced. 2) a geometry of array is unrestricted for solving the narrowband coherency problem. 3) the efficiency of signal decorrelation does not depend on the phase differences between coherent signals unlike the Coherent Signal Subspace Method (CSM). Simulation results are illustrated to demonstrate the superior performance of this new algorithm in comparison with the normal MUSIC and examine the comparative performance with the various choices of the optimal transformation matrix under coherent signal environments.

• Journal of the Korean Statistical Society
• /
• 제33권1호
• /
• pp.107-128
• /
• 2004

Bivariate Laplace distributions for which both marginal distributions and Laplace are discussed. Three kinds of bivariate Laplace distributions which are extended bivariate exponential distributions of Gumbel (1960) are introduced in this paper. These symmetrical distributions are compared with asymmetrical distributions of Kotz et al. (2000). Their probability density functions, cumulative distribution functions are derived. Conditional skewnesses and kurtoses are also defined. Their correlation coefficients are calculated and compared with others. We proposed bivariate random vector generating methods whose distributions are bivariate Laplace. With sample means and medians obtained from generated random vectors, variance and covariance matrices of means and medians are calculated and discussed with those of bivariate normal distribution.

• Journal of the Korean Data and Information Science Society
• /
• 제24권6호
• /
• pp.1455-1464
• /
• 2013

This paper develops maximum likelihood estimators (MLEs) of unknown parameters in an exponentiated half-logistic distribution based on a progressively type-II censored sample. We obtain approximate confidence intervals for the MLEs by using asymptotic variance and covariance matrices. Using importance sampling, we obtain Bayes estimators and corresponding credible intervals with the highest posterior density and Bayes predictive intervals for unknown parameters based on progressively type-II censored data from an exponentiated half logistic distribution. For illustration purposes, we examine the validity of the proposed estimation method by using real and simulated data.