• Communications for Statistical Applications and Methods
• /
• v.23 no.2
• /
• pp.105-117
• /
• 2016

In the paper, as a sequence of the first tutorial, we discuss sufficient dimension reduction methodologies used to estimate central subspace (sliced inverse regression, sliced average variance estimation), central mean subspace (ordinary least square, principal Hessian direction, iterative Hessian transformation), and central $k^{th}$-moment subspace (covariance method). Large-sample tests to determine the structural dimensions of the three target subspaces are well derived in most of the methodologies; however, a permutation test (which does not require large-sample distributions) is introduced. The test can be applied to the methodologies discussed in the paper. Theoretical relationships among the sufficient dimension reduction methodologies are also investigated and real data analysis is presented for illustration purposes. A seeded dimension reduction approach is then introduced for the methodologies to apply to large p small n regressions.

• Communications for Statistical Applications and Methods
• /
• v.23 no.2
• /
• pp.93-103
• /
• 2016

In the paper, we discuss dimension reduction of predictors ${\mathbf{X}}{\in}{{\mathbb{R}}^p}$ in a regression of $Y{\mid}{\mathbf{X}}$ with a notion of sufficiency that is called sufficient dimension reduction. In sufficient dimension reduction, the original predictors ${\mathbf{X}}$ are replaced by its lower-dimensional linear projection without loss of information on selected aspects of the conditional distribution. Depending on the aspects, the central subspace, the central mean subspace and the central $k^{th}$-moment subspace are defined and investigated as primary interests. Then the relationships among the three subspaces and the changes in the three subspaces for non-singular transformation of ${\mathbf{X}}$ are studied. We discuss the two conditions to guarantee the existence of the three subspaces that constrain the marginal distribution of ${\mathbf{X}}$ and the conditional distribution of $Y{\mid}{\mathbf{X}}$. A general approach to estimate them is also introduced along with an explanation for conditions commonly assumed in most sufficient dimension reduction methodologies.

• Communications for Statistical Applications and Methods
• /
• v.19 no.6
• /
• pp.885-898
• /
• 2012

In this paper, we consider Bayesian approaches to partially linear models, in which a regression function is represented by a semiparametric additive form of a parametric linear regression function and a nonparametric regression function. We make a comparative study on the performance of widely used Bayesian partially linear models in terms of empirical analysis. Specifically, we deal with three Bayesian methods to estimate the nonparametric regression function, one method using Fourier series representation, the other method based on Gaussian process regression approach, and the third method based on the smoothness of the function and differencing. We compare the numerical performance of three methods by the root mean squared error(RMSE). For empirical analysis, we consider synthetic data with simulation studies and real data application by fitting each of them with three Bayesian methods and comparing the RMSEs.

• Communications for Statistical Applications and Methods
• /
• v.21 no.2
• /
• pp.183-191
• /
• 2014

We define a multivariate Cauchy distribution using a probability density function; subsequently, a Ferguson's definition of a multivariate Cauchy distribution can be viewed as a characterization theorem using the characteristic function approach. To clarify this characterization theorem, we construct two dependent Cauchy random variables, but their sum is not Cauchy distributed. In doing so the proofs depend on the characteristic function, but we use the cumulative distribution function to obtain the exact density of their sum. The derivation methods are relatively straightforward and appropriate for graduate level statistics theory courses.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.1031-1036
• /
• 2009

Under recent dramatic declines in response rates, various procedures have been considered among survey practitioners to reduce nonresponse in order to avoid its potential impairment to the inference. In the random digit dialing telephone surveys, substantial efforts are often required to obtain the initial contact for the screener interview. To reduce a burden with higher data collection costs, refusal conversion can be administered only to a random portion of the sample, reducing nonresponse (bias) with an expense of sample variability increment due to the associated weight adjustment. In this paper, we provide ways to determine the optimal subsampling rate using a linear cost model. Our approach for refusal subsampling is to predetermine a random portion from the full sample and to apply refusal conversion efforts if needed only to the subsample.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.891-901
• /
• 2009

In this paper, we investigate the approach to estimate VaR under the transformed GARCH model. The time series are transformed to approximate to the underlying distribution of error terms and then the parameters and the one-sided prediction interval are estimated with the transformed data. The back-transformation is applied to compute the VaR in the original data scale. The analyses on the asset returns of KOSPI and KOSDAQ are presented to verify the accuracy of the coverage probabilities of the proposed VaR.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.971-978
• /
• 2009

Poisson generalized linear mixed models(GLMMs) have been widely used for the analysis of clustered or correlated count data. For the inference marginal likelihood, which is obtained by integrating out random effects is often used. It gives maximum likelihood(ML) estimator, but the integration is usually intractable. In this paper, we propose how to obtain the ML estimator via Laplace approximation based on hierarchical-likelihood (h-likelihood) approach under the Poisson GLMMs. In particular, the h-likelihood avoids the integration itself and gives a statistically efficient procedure for various random-effect models including GLMMs. The proposed method is illustrated using two practical examples and simulation studies.

• Communications for Statistical Applications and Methods
• /
• v.16 no.6
• /
• pp.997-1004
• /
• 2009

A standard approach is to classify all of future observations. In some cases, however, it would be desirable to defer a decision in particular for observations which are hard to classify. That is, it would be better to take more advanced tests rather than to make a decision right away. This motivates a classifier with a reject option that reports a warning for those observations that are hard to classify. In this paper, we present the method which gives efficient computation with a reject option. Some numerical results show strong potential of the propose method.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.10
• /
• pp.939-944
• /
• 2004

The numerical methods including the performance analysis and the noise prediction of the centrifugal compressor impeller are coupled with the optimization design skill, which consists of response surface method, statistical approach, and genetic algorithm. The flow-field Inside of a centrifugal compressor is obtained numerically by solving Wavier-Stokes equations. and then the propagating noise is estimated from the distributed surface pressure by using Ffowcs Williams-Hawkings formulation. The quadratic response surface model with D-optimal 3-level factorial experimental design points is constructed to optimize the impeller geometry for the advanced centrifugal compressor. The statistical analysis shows that the quadratic model exhibits a reasonable fitting quality resulting in the impeller blade design with high performance and low far-field noise level. The influences of selected design variables, objective functions, and constraints on the impeller performance and the impeller noise are also examined as a result of the optimization process.

• Korean Journal of Clinical Laboratory Science
• /
• v.40 no.1
• /
• pp.18-25
• /
• 2008

In classical titrimetric analyses, the major concern is the concentration of titrant, usually the aqueous solution of hydrochloric acid or sodium hydroxide, that could be changed as time goes by and it is accompanied with the inaccuracy of the resulting data. And the statistical approach, the nonlinear regression analysis, which is a well-known statistical method, was introduced to determine the accurate concentration of the titrant and the exact value of parameters, $K_a$, r, $C_a$, $C_b$, for 0.01 M aqueous solutions of analytes, sodium pyruvate, sodium acetate, sodium bicarbonate, ammonium hydroxide, ammonium chloride and acetic acid at $25^{\circ}C$. We used Gauss-Newton method for the linearlization of the nonlinear titration system and the two-parameter fitting showed appreciable convergent data for the parameters of the analytes set with the various range of $K_a$ value.