• Communications for Statistical Applications and Methods
• /
• v.5 no.3
• /
• pp.895-904
• /
• 1998

Two stage shrinkage testimator is a kind of adaptive estimators based on a test on an initial estimate of parameter. Since weighing factor plays an important roll in assessing the properties of testimator, its choice is extremely crucial in two stage testimation. Adke, Waikar and Schuurmann(1987) proposed a testimator for the mean of an exponential distribution defined with their own weighing factor. Two alternative testimators obtained using changed weighing factors are presented, and their Mean squared error(MSE) formulae are provided in this paper. Their properties are compared with those of existing one by means of MSE.

• Communications for Statistical Applications and Methods
• /
• v.20 no.4
• /
• pp.329-337
• /
• 2013

Consider a p-variate($p{\geq}3$) normal distribution with mean ${\theta}$ and covariance matrix ${\sum}={\sigma}^2{\mathbf{I}}_p$ for any unknown scalar ${\sigma}^2$. In this paper we improve the James-Stein estimator of ${\theta}$ in cases of shrinking toward some vectors using the Stein variance estimator. It is also shown that this domination does not hold for the positive part versions of these estimators.

• Journal of the Korean Statistical Society
• /
• v.31 no.3
• /
• pp.343-357
• /
• 2002

This paper proposes a method for producing smooth and positive estimates of the intensity function of an inhomogeneous Poisson process based on the shrinkage of wavelet coefficients of the observed counts. The information projection is used in conjunction with the level-dependent thresholds to yield smooth and positive estimates. This work is motivated by and demonstrated within the context of a problem involving gamma-ray burst data in astronomy. Simulation results are also presented in order to show the performance of the information projection estimators.

• Journal of the Korean Statistical Society
• /
• v.24 no.1
• /
• pp.257-266
• /
• 1995

Consider a p-variate $(p \geq 4)$ normal distribution with mean $\b{\theta}$ and identity covariance matrix. For estimating $\b{\theta}$ under a quadratic loss we investigate the behavior of risks of Stein-type estimators which shrink the usual estimator toward the mean of observations. By using concavity of the function appearing in the shrinkage factor together with new expectation identities for noncentral chi-squared random variables, a characterization of estimators with nondecreasing risk is obtained.

• The Korean Journal of Applied Statistics
• /
• v.21 no.1
• /
• pp.109-123
• /
• 2008

Many small area estimation methods have been suggested. Also for the comparison of the estimation methods, model diagnostic checking techniques have been studied. Almost all of the small area estimators were developed by minimizing MSE(Mean square error) and so the MSE is the well-known comparison criterion for superiority. In this paper we suggested a new small area estimator based on minimizing MSPE(Mean square percentage error) which is recently re-highlighted. Also we compared the new suggested estimator with the estimators explained in Shin et al. (2007) using MSE, MSPE and other diagnostic checking criteria.

• The Korean Journal of Applied Statistics
• /
• v.27 no.4
• /
• pp.605-617
• /
• 2014

Small area estimation is a statistical inference method to overcome large variance due to a small sample size allocated in a small area. A shrinkage estimator obtained by minimizing relative error(RE) instead of MSE has been suggested. The estimator takes advantage of good interpretation when the data range is large. A semiparametric estimator is also studied for small area estimation. In this study, we suggest a semiparametric shrinkage small area estimator and compare small area estimators using labor statistics.

• Journal of Communications and Networks
• /
• v.16 no.6
• /
• 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.

• The Korean Journal of Applied Statistics
• /
• v.7 no.1
• /
• pp.75-82
• /
• 1994

It is known that collinearity among the explanatory variables in generalized linear models inflates the variance of maximum likelihood estimators. A ridge-type estimator is presented using penalized likelihood. A method for choosing a shrinkage parameter is discussed and this method is based on a prediction-oriented criterion, which is Mallow's $C_L$ statistic in a linear regression setting.

• International Journal of Reliability and Applications
• /
• v.7 no.2
• /
• pp.167-175
• /
• 2006

Effective and quick methods that are easy to carry out even by hand, or easy to be programmed by hand-held calculators are needed for assessing the sizes of contrasts of unreplicated $2^P$ factorial designs. Moreover, they have the advantage to use the original numerical measurements which makes the analysis easier to explain. Basically, Lenth (1989) is one of the most familiar of such quick and powerful methods. Later on, Aboukalam (2001) proposes under constant effects an alternative sophisticated method to Lenth's method. The proposed method is the supreme from two considerable powers. The first utmost indicates less inflation deficiency while the other utmost indicates less shrinkage deficiency. Also under constant effects, Al-Shiha (2006) introduces an alternative quick method which is less shrinkage deficiency while the inflation deficiency is the same. If effects are random, Aboukalam (2005) introduces an alternative quick method in which the first power is favored as long as the second power is within a small margin. In the spirit of quickness and fixed effects, this article adds another method which is supreme from the two considerable powers. The method is based on a one step of the scale-part of a suggested M-estimate for location. Explicitly, we suggest adapting the skipped median (ASKM) estimate. Critical values of ASKM-method, for several sample sizes often used, are empirically computed.

• Communications for Statistical Applications and Methods
• /
• v.25 no.2
• /
• pp.235-243
• /
• 2018

Many studies exist on the influence of one or few observations on estimators in a variety of statistical models under the "large n, small p" setup; however, diagnostic issues in the regression models have been rarely studied in a high dimensional setup. In the high dimensional data, the influence of observations is more serious because the sample size n is significantly less than the number variables p. Here, we investigate the influence of observations on the least absolute shrinkage and selection operator (LASSO) estimates, suggested by Tibshirani (Journal of the Royal Statistical Society, Series B, 73, 273-282, 1996), and the influence of observations on selected variables by the LASSO in the high dimensional setup. We also derived an analytic expression for the influence of the k observation on LASSO estimates in simple linear regression. Numerical studies based on artificial data and real data are done for illustration. Numerical results showed that the influence of observations on the LASSO estimates and the selected variables by the LASSO in the high dimensional setup is more severe than that in the usual "large n, small p" setup.