• Communications for Statistical Applications and Methods
• /
• 제12권3호
• /
• pp.615-624
• /
• 2005

In this paper, we review the variable-selection properties of LASSO and SCAD in penalized regression. To improve the weakness of SCAD for high noise level, we propose a new penalty function called MSCAD which relaxes the unbiasedness condition of SCAD. In order to compare MSCAD with LASSO and SCAD, comparative studies are performed on simulated datasets and also on a real dataset. The performances of penalized regression methods are compared in terms of relative model error and the estimates of coefficients. The results of experiments show that the performance of MSCAD is between those of LASSO and SCAD as expected.

• Communications for Statistical Applications and Methods
• /
• 제14권2호
• /
• pp.287-299
• /
• 2007

We consider a multiclass queue where queued customers are served in their order of arrival at a rate which depends on the customer type. By using the asymptotic results obtained by Dabrowski et al. (2006) we calculate the sharp asymptotics of the stationary distribution of the number of customers of each class in the system and the distribution of the number of customers of each class when the total number of customers reaches a high level before emptying. We also obtain a fast simulation algorithm to estimate the overflow probability and compare it with the general simulation and asymptotic results.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2005년도 Proceedings of ISRS 2005
• /
• pp.26-29
• /
• 2005

In order to compute cloud coverage statistics over Asian region, an operational scheme for masking cloud-contaminated pixels in Advanced Very High Resolution Radiometer (AVHRR) daytime data was developed, evaluated and presented. Dynamic thresholding was used with channell, 2 and 3 to automatically create a cloud mask for a single image. Then the IO-day cloud coverage imagery was generated over the whole Asian region along with cloud-free composite imagery. Finally the monthly based statistics were computed based on the derived cloud coverage imagery in terms of land cover and country. As a result, it was found that 20-day is required to acquire the cloud free data over the whole Asia using NOAA AVHRR. The to-day cloud coverage and cloud-free composite imagery derived in this research is available via the web-site http://webpanda.iis.u-tokyo.ac.jp/CloudCover/.

• Journal of the Korean Statistical Society
• /
• 제30권4호
• /
• pp.667-678
• /
• 2001

The asymptotic behavior and distribution for quantiles estimators using ranked samples are introduced. Applications of quantiles estimation on finding the normal ranges (2.5% and 97.5% percentiles) and the median of some medical characteristics and on finding the Hodges-Lehmann estimate are discussed. The conclusion of this study is, whenever perfect ranking is possible, the relative efficiency of quantiles estimation using ranked samples relative to SRS is high. This may translates to large savings in cost and time. Also, this conclusion holds even if the ranking is not perfect. Computer simulation results are given and real data from lows 65+ study is used to illustrate the method.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2005년도 춘계 학술발표회 논문집
• /
• pp.7-12
• /
• 2005

In this paper, we review the variable-selection properties of LASSO and SCAD in penalized regression. To improve the weakness of SCAD for high noise level, we propose a new penalty function called MSCAD which relaxes the unbiasedness condition of SCAD. In order to compare MSCAD with LASSO and SCAD, comparative studies are performed on simulated datasets and also on a real dataset. The performances of penalized regression methods are compared in terms of relative model error and the estimates of coefficients. The results of experiments show that the performance of MSCAD is between those of LASSO and SCAD as expected.

• 대한안전경영과학회지
• /
• 제9권2호
• /
• pp.123-134
• /
• 2007

This paper focuses on the in-house reliability assurance plan for the bulk materials of each company. The reliability assurance needs in essence a long time and high cost for testing the materials. In order to reduce the time and cost, accelerated life test is adopted. The bulk sampling technique was used for acceptance. Design parameters might be total sample size(segments and increments}, stress level and so on. We focus on deciding the sample size by minimizing the asymptotic variance of test statistics as well as satisfying the consumer's risk. In bulk sampling, we also induce the sample size by adapting the normal life time distribution model when the variable of the lognormal life time distribution is transformed and adapted to the model. In addition, the sample size for both the segments and increments can be induced by minimizing the asymptotic variance of test statistics of the segments and increments with consumer's risk met. We can assure the reliability of the mean life and B100p life time of the bulk materials by using the calculated minimum sample size.

• 응용통계연구
• /
• 제18권3호
• /
• pp.713-727
• /
• 2005

인터넷 조사는 전통적인 방법에 비해 자료수집이 신속하고, 조사비용이 저렴하며, 양질의 자료를 얻을 수 있으며 멀티미디어를 활용한 고도화된 설문을 설계할 수 있다. 그러나 조사대상자의 신원 파악이 어려우며 구체적인 고객만족도에 대한 분석이 어려운 상황이다. 따라서 인터넷을 이용한 설문조사뿐만 아니라 고객만족도 조사까지의 확장이 가능한 시스템을 개발하였다. 이 시스템은 조사계획 과정에서 미리 조사 대상자에 대한 기본 정보를 취합하여 E-mail을 통해 조사 대상자의 특성을 파악할 수 있다. 또한 시스템을 통한 계획의 입력과 진행의 현황을 확인할 수 있다.

• 응용통계연구
• /
• 제32권6호
• /
• pp.909-922
• /
• 2019

벌점화 추정 기법 중 adaptive LASSO 방법은 모형 선택과 모수 추정을 동시에 할 수 있는 유명한 방법으로 이미 정상 자기회귀모형에서 연구된 적이 있다. 본 논문에서는 이를 확장하여 확률보행과정과 같은 비정상 자기회귀모형에서 adaptive LASSO 추정량이 갖는 성질을 모의실험을 통해 연구하였다. 다만 비정상 자기회귀모형에서는 단위근의 존재 여부를 판단하는 것과 모형의 차수를 선택하는 것이 가장 중요하므로, 이를 위해 원 자기회귀모형이 아닌 ADF 검정에서 고려하는 회귀모형으로 변환하여 adaptive LASSO를 적용하였다. 일반적으로 Adaptive LASSO를 적용할 때 조절모수의 선택이 가장 중요한 문제이며, 본 논문에서는 교차검증, AIC, BIC 세 가지 방법을 이용하여 조절모수를 선택하였다. 모의실험 결과를 보면, 이 중에서 BIC가 최소가 되도록 선택한 조절모수에 대응되는 adaptive LASSO 추정량이 단위근의 존재 여부를 잘 판단할 뿐만 아니라 자기회귀모형의 차수 또한 비교적 정확하게 선택함을 확인할 수 있다.

• 한국통신학회논문지
• /
• 제31권10C호
• /
• pp.973-980
• /
• 2006

이제까지 국소 최적 검파를 다룬 연구들에서는 대부분 관측을 독립이라 두었다. 독립 관측 모형에서 얻은 검파기는 의존성 잡음 성분이 있는 현대 고속 통신 시스템에서 성능이 쾌 떨어질 수 있다. 이 논문의 1부에서는 곱셈 꼴 잡음과 일차 마르코프 덧셈꼴 잡음이 일어나는 환경에서 알려진 약한 신호를 검파할 때 알맞은 검정 통계량을 자세히 살펴본다. 이어, 2부에서는 여러 검파기의 점근 성능과 유한 표본 크기 성능을 얻고 서로 견주어 보며, 성능을 가장 좋게 하려면 간섭끼리의 의존성을 생각하여 검파기를 꾸며야 함을 보인다.

• 품질경영학회지
• /
• 제44권3호
• /
• pp.677-688
• /
• 2016

Purpose: In this study, we examine the effects of circulatory diseases mortality in South Korea 2005-2013 using the air pollution index, Methods: We cluster the region of high risk mortality by SaTScan$^{TM}$9.3.1 and compare this result with the regional distribution of air pollution. We use the Geographically Weighted Regression (GWR) to consider the spatial heterogeneity of data collected by administrative district in order to estimate the model. As GWR is spatial analysis techniques utilizing the spatial information, regression model estimated for each region on the assumption that regression coefficients are different by region. Results: As a result of estimating model of the collected air pollution index, circulatory diseases mortality data combined with the spatial information, GWR was found to solve the problem of spatial autocorrelation and increase the fit of the model than OLS regression model. Conclusion: GWR is used to select the air pollution affecting the disease each year, the K-means cluster analysis discover the characteristics of the distribution of air pollution by region.