• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1992.04b
• /
• pp.127-136
• /
• 1992

시뮬레이션 실험에서 시스템 성과에 대한 추정치의 정확도를 개선하기 위한 분산감소기법(Variance Reduction Technique)은 입력영역과 출력영역에 대한 것으로 나누어 볼 수 있다. 본 연구에서는 시스템 성과 추정량이 단일 변량인 경우에, 분산감소기법으로 많이 사용되는 Antithetic Variates방법과 Control Variates방법을 결합하여 응용가능한 시뮬레이션 실험설계기법을 제시하고 이 기법을 선택된 모형에 적용하여 시뮬레이션의 효율성을 분석하였다. 실험결과, 제안된 기법은 기존 방법들보다 추정치의 분산을 5%-8% 더 감소시켰다. 비록 제한된 실험결과이지만 이러한 효과는 대형 시뮬레이션의 경우에 적지 않으리라 기대된다. 특히 효과적인 Control Variates의 수가 적은 경우, 제안된 기법은 매우 효율적이다.

• Industrial Engineering and Management Systems
• /
• v.13 no.4
• /
• pp.442-448
• /
• 2014

Portfolio optimization in the presence of estimation error can be stabilized by incorporating norm-constraints; this result was shown by DeMiguel et al. (A generalized approach to portfolio optimization: improving performance by constraining portfolio norms, Management Science, 5, 798-812, 2009), who reported empirical performance better than numerous competing approaches. We extend the idea of norm-constraints by introducing a powerful enhancement, grouped selection for portfolio optimization. Here, instead of merely penalizing norms of the assets being selected, we penalize groups, where within a group assets are treated alike, but across groups, the penalization may differ. The idea of groupwise selection is grounded in statistics, but to our knowledge, it is novel in the context of portfolio optimization. Novelty aside, the real benefits of groupwise selection are substantiated by experiments; our results show that groupwise asset selection leads to strategies with lower variance, higher Sharpe ratios, and even higher expected returns than the ordinary norm-constrained formulations.

• Journal of Biomedical Engineering Research
• /
• v.15 no.1
• /
• pp.105-110
• /
• 1994

In this paper, we present a method to evaluate source locations and distributed region which is specified brain activity, as indicated by locations and strengths of intracranial sources, using potential gradients of interpolation polynomials and topographic mapping of the EEG records. This method can analyze the variance of source temporally or spatially and leads to enable a quantitative evaluation of potential gradients drawing methods which is now being used in the clinic. In the result, we obtained the overall potentials distribution on the entire scalp and the information of potential source locations from the EEG records of a patient which was known to epilepsy.

• Communications for Statistical Applications and Methods
• /
• v.6 no.2
• /
• pp.443-456
• /
• 1999

The estimation of mean lifetimes in presence of interval censoring with mixed replacement procedure are examined when the distribution s of lifetimes are exponential. it is assumed that due to physical restrictions and/or economic constraints the number of failures is investigated only at several inspection times during the lifetime test; thus there is interval censoring. Comparisons of mixed replacement designs are made with those with and without replacement The maximum likelihood estimator is found in an implicit form. The Cramer-Rao lower bound which is the asymptotic variance of the estimator is derived. The test conditions for minimizing the Cramer-Rao lower bound and minimizing the test costs within a desired width of the Cramer-Rao bound have been studied.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2000.05a
• /
• pp.367-370
• /
• 2000

Various methods of enhancing the performance of passive range sonar arrays have been discussed, triangulation, wavefront curvature method etc. But they are not appropriate to the methods because of very low SNR in underwater environment. We made appropriate sub-arrays in a linear array and applied to the beamformers such as a minimum variance with null constraints.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2006.12a
• /
• pp.67-83
• /
• 2006

In successive sampling on two occasions the problem of estimating a finite population quantile has been considered. The theory developed aims at providing the optimum estimates by combining (i) three double sampling estimators viz. ratio-type, product-type and regression-type, from the matched portion of the sample and (ii) a simple quantile based on a random sample from the unmatched portion of the sample on the second occasion. The approximate variance formulae of the suggested estimators have been obtained. Optimal matching fraction is discussed. A simulation study is carried out in order to compare the three estimators and direct estimator. It is found that the performance of the regression-type estimator is the best among all the estimators discussed here.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.6
• /
• pp.372-372
• /
• 2000

In this paper, we propose to use nonparametric detector for the PN code acquisition in DS/CDMA systems. Because the detector is nonparametric, we can decide the threshold of the detector without first haying to estimate the valiance of time-varying interference. We compare the proposed detector with the conventional detector by showing both analytic and simulation results. From these results, we can observe that the DS/CDMA system with the proposed detector have better performance than that with the conventional detector when the exact estimation of the interference variance is not possible.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.05a
• /
• pp.1085-1092
• /
• 2006

Response surface methodology(RSM) is a useful collection of experimentation techniques for developing, improving, and optimizing products and processes. When we are to estimate second-order regression model and optimize quality characteristic by RSM, central composite designs and Box-Behnken designs are widely in use. However, in developing cutting-edge products, it is very crucial to reduce the time of experimentation as much as possible. In this paper small-sized second-order designs are introduced and their estimation abilities are compared in terms of D-optimality, A-optimality, and variance of regression coefficients. The result of this study will be beneficial to experimenters who face experiments which are expensive, difficult, or time-consuming.

• The Journal of the Acoustical Society of Korea
• /
• v.27 no.1E
• /
• pp.1-10
• /
• 2008

A noise suppression algorithm is proposed for nonstationary noisy environments. The proposed algorithm is different from the conventional approaches such as the spectral subtraction algorithm and the minimum statistics noise estimation algorithm in that it classifies speech and noise signals in time-frequency bins. It calculates the ratio of the variance of the noisy power spectrum in time-frequency bins to its normalized time-frequency average. If the ratio is greater than an adaptive threshold, speech is considered to be present. Our adaptive algorithm tracks the threshold and controls the trade-off between residual noise and distortion. The estimated clean speech power spectrum is obtained by a modified gain function and the updated noisy power spectrum of the time-frequency bin. This new algorithm has the advantages of simplicity and light computational load for estimating the noise. This algorithm reduces the residual noise significantly, and is superior to the conventional methods.

• Honam Mathematical Journal
• /
• v.25 no.1
• /
• pp.95-115
• /
• 2003

Consider the problem of estimating a $p{\times}1$ mean vector ${\theta}(p{\geq}4)$ under the quadratic loss, based on a sample $X_1,\;{\cdots}X_n$. We find an optimal decision rule within the class of Lindley type decision rules which shrink the usual one toward the mean of observations when the underlying distribution is that of a variance mixture of normals and when the norm $||{\theta}-{\bar{\theta}}1||$ is known, where ${\bar{\theta}}=(1/p)\sum_{i=1}^p{\theta}_i$ and 1 is the column vector of ones. When the norm is restricted to a known interval, typically no optimal Lindley type rule exists but we characterize a minimal complete class within the class of Lindley type decision rules. We also characterize the subclass of Lindley type decision rules that dominate the sample mean.