• Communications for Statistical Applications and Methods
• /
• 제2권1호
• /
• pp.145-154
• /
• 1995

Empirical Bayes approach is applied to estimation of the binomial parameter when there is a cost for observations. Both the sample size and the decision rule for estimating the parameter are determined stochastically by the data, making the result more useful in applications. Our empirical Bayes problems with non-iid components are compared to the usual empirical Bayes problems with iid components. The asymptotic optimal procedure with a computer simulation is given.

• Journal of the Korean Statistical Society
• /
• 제20권1호
• /
• pp.67-76
• /
• 1991

The empirical Bayes version involves ″independent″ repetitions(a sequence) of the component decision problem. With the varying sample size possible, these are not identical components. However, we impose the usual assumption that the parameters sequence $\theta$=($\theta$$_1$, $\theta$$_2$, …) consists of independent G-distributed parameters where G is unknown. We assume that G $\in$ g, a known family of distributions. The sample size $N_i$ and the decisin rule $d_i$ for component i of the sequence are determined in an evolutionary way. The sample size $N_1$ and the decision rule $d_1$$\in$ $D_{N1}$ used in the first component are fixed and chosen in advance. The sample size $N_2$and the decision rule $d_2$ are functions of *see full text($\underline{X}^1$equation omitted), the observations in the first component. In general, $N_i$ is an integer-valued function of *see full text(equation omitted) and, given $N_i$, $d_i$ is a $D_{Ni}$/-valued function of *see full text(equation omitted). The action chosen in the i-th component is *(equation omitted) which hides the display of dependence on *(equation omitted). We construct an empirical Bayes decision rule for estimating normal mean and show that it is asymptotically optimal.

• Journal of the Korean Statistical Society
• /
• 제11권1호
• /
• pp.25-35
• /
• 1982

A problem of selecting the least probable cell in a multinomial distribution is studied in a Bayesian framework. We consider two loss components the cost of sampling and the difference in cell probabilities between the selected and the least probable cells. A Bayes sequential selection rule is derived with respect to a Dirichlet prior, and it is compared with the best fixed sample size selection rule. The continuation sets with respect to the vague prior are tabulated for certain cases.

• 품질경영학회지
• /
• 제22권3호
• /
• pp.179-190
• /
• 1994

Financial variables have been used in bankruptcy prediction. Despite of possible errors in prediction, most existing approaches do not consider the causal time sequence of prediction activity and bankruptcy phenomena. This paper proposes a prediction method using Neyman-Pearson Theorem and Bayes' rule. The proposed method uses posterior probability concept and determines a prediction policy with appropriate error rate.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 D
• /
• pp.2474-2476
• /
• 2004

지폐 인식 자동화기기가 여러 분야에 보편화되면서 다양한 지폐를 고속으로 처리할 수 있는 고속지폐 인식 자동화 기기가 요구되고 있다. 하지만 대부분의 지폐 인식 자동화 기기가 고속화에 적합하지 않은 구조로 설계되어 있고 신권 추가가 용이하지 않다. 본 논문은 고속 Paper Currency 인식 시스템에 적합한 범용 하드웨어 시스템과 Bayes Rule 기반의 고속 인식 알고리즘을 제안한다. 제안된 범용 하드웨어 구조는 고속의 CIS(Contact Image Sensor)와 DSP(Digital Signal Processor) 그리고 Dual Memory System으로 구성되었다. Bayes Rule에 기초한 고속 인식 알고리즘은 기존의 Paper Currency 인식 시스템에 사용되었던 기계학습 방법에 비해 신권 추가가 쉽고 적은 연산으로 권종을 판별할 수 있어 고속 지폐 인식 자동화기기에 적합하다. 본 논문에서는 제안된 방법들을 실제 자동화기기로 구현하여 그 유용성을 검증한다.

• 한국정밀공학회지
• /
• 제17권7호
• /
• pp.98-105
• /
• 2000

In this study, we developed rotating machine vibration condition monitoring system based upon Windows NT and DSP Board. Developed system includes signal analysis module, trend monitoring and simple diagnosis using threshold value. Trend analysis and report generation are offered with database management tool which was developed in MS-ACCESS environment. Post-processor, based upon Matlab, is developed for vibration signal analysis and fault detection using statistical pattern recognition scheme based upon Bayes discrimination rule and neural networks. Concerning to Bayes discrimination rule, the developed system contains the linear discrimination rule with common covariance matrices and the quadratic discrimination rule under different covariance matrices. Also the system contains k-nearest neighbor method to directly estimate a posterior probability of each class. The result of case studies with the data acquired from Pyung-tak LNG pump and experimental setup show that the system developed in this research is very effective and useful.

• 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.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1606-1609
• /
• 2002

MAC state models appeared with an effort to overcome technical demerits of CDMA in provisioning packet data service. In the scenario of sojourn and transition on MAC states, the design of state sojourn time is a critical issue for an efficient utilization of limited recource; a longer sojourn time leads to more resource being preserved for inactive stations, while more connection components should be recovered with a shorter sojourn time. Thus, the sojourn time at each MAC state must be optimized in consideration of these two conflicting arguments. In this paper, we first present a generic MAC state model. Secondly, based on the generic model, we reveal a relation of inactive period and the delay time of the last packet served in pre- ceding active period and specify a loss function reflect-ing two antinomic features that result from a change of state sojourn time. Using the proposed loss function, we construct a decision problem to find an optima3 rule for state sojourn times. Finally, we present a way of computing Bayes rule by use of the posterior distribution of inactivity duration for given observation on the delay time of last packet. Furthermore, Bayes rules are explicitly expressed for special arrival processes and investigated with respect to traffic load and loss parameters.

• 한국정밀공학회지
• /
• 제16권12호
• /
• pp.119-125
• /
• 1999

This paper gives an overview of the various approaches to designing statistical pattern recognition scheme based on Bayes discrimination rule and the artificial neural networks for rotating machine condition classification. Concerning to Bayes discrimination rule, this paper contains the linear discrimination rule applied to classification into several multivariate normal distributions with common covariance matrices, the quadratic discrimination rule under different covariance matrices. Also we discribes k-nearest neighbor method to directly estimate a posterior probability of each class. Five features are extracted in time domain vibration signals. Employing these five features, statistical pattern classifier and neural networks have been established to detect defects on rotating machine. Four different cases of rotation machine were observed. The effects of k number and neural networks structures on monitoring performance have also been investigated. For the comparison of diagnosis performance of these two method, their recognition success rates are calculated form the test data. The result of experiment which classifies the rotating machine conditions using each method presents that the neural networks shows the highest recognition rate.

• 품질경영학회지
• /
• 제29권1호
• /
• pp.11-23
• /
• 2001

This paper is concerned with suggesting a Bayesian method for variable selection in multinomial logit model. It is based upon an optimal rule suggested by use of Bayes rule which minimizes a risk induced by selecting the multinomial logit model. The rule is to find a subset of variables that maximizes the marginal likelihood of the model. We also propose a Laplace-Metropolis algorithm intended to suggest a simple method forestimating the marginal likelihood of the model. Based upon two examples, artificial data and empirical data examples, the Bayesian method is illustrated and its efficiency is examined.