• East Asian mathematical journal
• /
• 제5권1호
• /
• pp.95-110
• /
• 1989

In a general variance component model, nonnegative quadratic estimators of the components of variance are considered which are invariant with respect to mean value translaion and have minimum bias (analogously to estimation theory of mean value parameters). Here the minimum is taken over an appropriate cone of positive semidefinite matrices, after having made a reduction by invariance. Among these estimators, which always exist the one of minimum norm is characterized. This characterization is achieved by systems of necessary and sufficient condition, and by a cone restricted pseudoinverse. In models where the decomposing covariance matrices span a commutative quadratic subspace, a representation of the considered estimator is derived that requires merely to solve an ordinary convex quadratic optimization problem. As an example, we present the two way nested classification random model. An unbiased estimator is derived for the mean squared error of any unbiased or biased estimator that is expressible as a linear combination of independent sums of squares. Further, it is shown that, for the classical balanced variance component models, this estimator is the best invariant unbiased estimator, for the variance of the ANOVA estimator and for the mean squared error of the nonnegative minimum biased estimator. As an example, the balanced two way nested classification model with ramdom effects if considered.

• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1169-1180
• /
• 2006

Partial least squares regression (PLS) is a biased, non-least squares regression method and is an alternative to the ordinary least squares regression (OLS) when predictors are highly collinear or predictors outnumber observations. One way to understand the properties of biased regression methods is to know how the estimators shrink the OLS estimator. In this paper, we introduce an expression for the shrinkage factor of PLS and develop a new shrinkage expression, and then prove the equivalence of the two representations. We use two near-infrared (NIR) data sets to show general behavior of the shrinkage and in particular for what eigendirections PLS expands the OLS coefficients.

• Communications for Statistical Applications and Methods
• /
• 제14권3호
• /
• pp.641-648
• /
• 2007

We propose the new sampling method, called modified ranked ordering set sampling (MROSS). Kim and Kim (2003) suggested the sign test using the ranked ordering set sampling (ROSS), and showed that the asymptotic relative efficiency (ARE) of ROSS against RSS for sign test increases as sample size does. We propose the estimator for the population mean using MROSS. The relative precision (RP) of estimator of the population mean using MROSS method with respect to the usual estimator using modified RSS is higher, and when the underlying distribution is skewed, the bias of the proposed estimator is smaller than that of several ranked set sampling estimators.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제12권3호
• /
• pp.153-160
• /
• 2008

A size-biased Poisson distribution is defined. Its characterization by using a recurrence relation for first order negative moment of the distribution is obtained. Different estimation methods for the parameter of the model are also discussed. R-Software has been used for making a comparison among the three different estimation methods.

• Journal of the Korean Statistical Society
• /
• 제12권2호
• /
• pp.81-90
• /
• 1983

The least squares estimator of disturbance variance in a regression model is biased under a serial correlation. Under the assumption of an AR(I), Theil(1971) crudely related the bias with the autocorrelation of the disturbances and the autocorrelation of the explanatory variable for a simple regression. In this paper we derive a relation which relates the bias with the autocorrelation of disturbances and the autocorrelation of explanatory variables for a multiple regression with improved precision.

• Journal of Communications and Networks
• /
• 제16권6호
• /
• pp.583-591
• /
• 2014

In wireless communications, knowledge of the signal-to-noise ratio is required in diverse communication applications. In this paper, we derive the variance of the maximum likelihood estimator in the data-aided and non-data-aided schemes for determining the optimal shrinkage factor. The shrinkage factor is usually the constant that is multiplied by the unbiased estimate and it increases the bias slightly while considerably decreasing the variance so that the overall mean squared error decreases. The closed-form biased estimators for binary-phase-shift-keying and quadrature phase-shift-keying systems are then obtained. Simulation results show that the mean squared error of the proposed method is lower than that of the maximum likelihood method for low and moderate signal-to-noise ratio conditions.

• 응용통계연구
• /
• 제31권3호
• /
• pp.397-408
• /
• 2018

표본조사에서는 정확한 모수 추정을 위한 다양한 방법이 개발되었으며 이 중에서 보조정보를 이용한 비추정량 또는 회귀추정량이 흔히 사용된다. 최근 많은 연구가 진행되고 있는 비형태추정량(ratio type estimator)은 비추정량의 단점을 보완하여 추정의 정확성을 향상시키는 것으로 알려져 있다. 그러나 비형태추정량은 편향이 있는 것으로 알려져 있어 이를 해결하기 위한 연구가 활발히 진행되고 있다. 이에 본 연구에서는 편향을 제거하기 위해 비형태추정량에 새로운 모수를 추가한 일반화 비형태추정량(generalized ratio-type estimator)을 제안하였다. 또한 사업체조사와 같이 등분산성을 만족하지 않는 자료에서 추정의 정확성 향상을 위해 모형의 오차에 포함된 분산 모수를 추정하고 제안된 추정량을 적용하는 방법을 제안하였다. 또한 모의실험을 통해 일반화 비형태추정량은 기존의 비추정량에 비해 매우 우수한 결과를 주는 것을 확인하였다.

• Smart Structures and Systems
• /
• 제17권4호
• /
• pp.647-667
• /
• 2016

In this paper, dynamic displacement is estimated with high accuracy by blending high-sampling rate acceleration data with low-sampling rate displacement measurement using a two-stage Kalman estimator. In Stage 1, the two-stage Kalman estimator first approximates dynamic displacement. Then, the estimator in Stage 2 estimates a bias with high accuracy and refines the displacement estimate from Stage 1. In the previous Kalman filter based displacement techniques, the estimation accuracy can deteriorate due to (1) the discontinuities produced when the estimate is adjusted by displacement measurement and (2) slow convergence at the beginning of estimation. To resolve these drawbacks, the previous techniques adopt smoothing techniques, which involve additional future measurements in the estimation. However, the smoothing techniques require more computational time and resources and hamper real-time estimation. The proposed technique addresses the drawbacks of the previous techniques without smoothing. The performance of the proposed technique is verified under various dynamic loading, sampling rate and noise level conditions via a series of numerical simulations and experiments. Its performance is also compared with those of the existing Kalman filter based techniques.

• 산업공학
• /
• 제22권3호
• /
• pp.214-222
• /
• 2009

In the early design stage, system reliability must be estimated from life testing data at the component level. Previously, a point estimate of system reliability was obtained from the unbiased estimate of the component reliability after assuming that the number of failed components for a given time followed a binomial distribution. For deriving the confidence interval of system reliability, either the lognormal distribution or the normal approximation of the binomial distribution was assumed for the estimator of system reliability. In this paper, a new estimator is used for the component level reliability, which is biased but has a smaller mean square error than the previous one. We propose to use the beta distribution rather than the lognormal or approximated normal distribution for developing the confidence interval of the system reliability. A numerical example based on Monte Carlo simulation illustrates advantages of the proposed approach over the previous approach.

• Journal of the Korean Data and Information Science Society
• /
• 제24권4호
• /
• pp.815-824
• /
• 2013

국가단위의 조사와 같은 대규모 표본조사에서는 표본의 대표성을 확보하기 위해 층화, 집락, 계통, 불균등확률추출 등을 종합적으로 사용하는 복합표본설계가 일반화되어 있다. 이러한 복합표본설계에 기초한 범주형 자료분석에서는 자료의 독립성과 다항분포를 가정하는 전통적인 피어슨검정이 왜곡된 검정결과를 가져올 수 있다. 본 연구는 복합표본설계에 의한 범주형조사자료의 k-모집단 동질성검정에서 설계기반 일치통계량인 Wald 검정통계량을 유도하고, 전통적인 피어슨검정통계량을 사용할 경우 발생할 수 있는 오차요인을 항목별로 분해하여, 분산의 편의에 의한 영향, 추정량의 편의에 의한 영향, 기타 분산의 편의와 추정량의 편의가 교락되어 미치는 영향으로 각각 분해하는 식을 도출하였다. 또한, 도출된 식의 각 항목이 피어슨 카이제곱검정통계량에 미치는 상대적 크기를 경험적으로 확인하기 위해 국민건강영양조사 제4기 2차년도 자료를 이용해 경험분석 하였다. 분석결과, 변수에 따른 차이는 있지만 대체로 분산의 편의가 미치는 영향이 추정량의 편의가 미치는 영향보다 크다는 것을 명확히 확인할 수 있었다.