• Journal of the Korean Data and Information Science Society
• /
• 제28권5호
• /
• pp.1191-1204
• /
• 2017

Weighted least squares (WLS) estimation is often easily used for the data with heteroscedastic errors because it is intuitive and computationally inexpensive. However, WLS estimator is less robust to a few outliers and sometimes it may be inefficient. In order to overcome robustness problems, Box-Cox transformation, Huber's M estimation, bisquare estimation, and Yohai's MM estimation have been proposed. Also, more efficient estimations than WLS have been suggested such as Bayesian methods (Cepeda and Achcar, 2009) and semiparametric methods (Kim and Ma, 2012) in heteroscedastic error models. Recently, Çelik (2015) proposed the weight methods applicable to the heteroscedasticity patterns including butterfly-distributed residuals and megaphone-shaped residuals. In this paper, we review heteroscedastic regression estimators related to robust or efficient estimation and describe their properties. Also, we analyze cost data of U.S. Electricity Producers in 1955 using the methods discussed in the paper.

• Communications for Statistical Applications and Methods
• /
• 제15권6호
• /
• pp.1003-1011
• /
• 2008

Mixed-effect Poisson regression models are widely used for analysis of correlated count data such as those found in longitudinal studies. In this paper, we consider kernel extensions with semiparametric fixed effects and parametric random effects. The estimation is through the penalized likelihood method based on kernel trick and our focus is on the efficient computation and the effective hyperparameter selection. For the selection of hyperparameters, cross-validation techniques are employed. Examples illustrating usage and features of the proposed method are provided.

• 응용통계연구
• /
• 제33권1호
• /
• pp.25-46
• /
• 2020

본 논문에서는 Bayesian spectral analysis regression (BSAR) 방법론을 이용한 베이지안 순서형 프로빗 준모수 회귀모형에 대해서 고찰한다. 순서형 프로빗 회귀모형은 순서가 있는 범주형 자료를 모형화하는 방법으로, 정규 분포의 분포함수의 역함수인 프로빗 연결함수를 이용해 각 범주의 확률과 설명변수을 연결함으로써 반응변수의 확률을 모형화한다. 베이지안 프로빗 회귀 모형은 정규 분포를 따르는 잠재변수를 도입함으로써 사후 분포 도출을 용이하게 하고, 절단점에 따라 나뉘어지는 잠재변수들의 값에 따라서 반응 변수들이 범주화된다. 본 논문에서는 이러한 잠재 변수 방법을 확장해 BSAR 방법론에 기반하여 단조증가/감소와 같은 형태제약을 반영할 수 있는 베이지안 이항형 및 순서형 프로빗 준모수 회귀모형에 대해 연구한다. 모의실험을 통하여 이항형 프로빗 준모수 회귀모형과 기존의 다른 모형들 간의 적합결과를 비교하고, 형태 제약에 따른 순서형 프로빗 준모수 회귀모형의 적합결과를 비교 분석하도록 한다. 아울러, 국민건강영양조사 제 7기 1차년도 (2016) 자료(Korean National Health and Nutrition Examination Survey (KNHANES), 2016)를 바탕으로, 본 논문에서 고찰한 이항형 및 순서형 프로빗 준모수 회귀모형을 적용하여, 흡연양태와 커피섭취 간의 관계에 대한 실증적 분석을 수행한다.

• Journal of the Korean Data and Information Science Society
• /
• 제21권4호
• /
• pp.765-775
• /
• 2010

For the accelerated failure time (AFT) model a lot of effort has been devoted to develop effective estimation methods. AFT model assumes a linear relationship between the logarithm of event time and covariates. In this paper we propose a semiparametric support vector machine to consider situations where the functional form of the effect of one or more covariates is unknown. The proposed estimating equation can be computed by a quadratic programming and a linear equation. We study the effect of several covariates on a censored response variable with an unknown probability distribution. We also provide a generalized approximate cross-validation method for choosing the hyper-parameters which affect the performance of the proposed approach. The proposed method is evaluated through simulations using the artificial example.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 추계 학술발표회 논문집
• /
• pp.179-182
• /
• 2003

This study describes a new graphical method for assessing and characterizing effect modification by a matching covariate in matched case-control studies. This method to understand effect modification is based on a semiparametric model using a varying coefficient model. The method allows for nonparametric relationships between effect modification and other covariates, or can be useful in suggesting parametric models. This method can be applied to examining effect modification by any ordered categorical or continuous covariates for which cases have been matched with controls. The method applies to effect modification when causality might be reasonably assumed. An example from veterinary medicine is used to demonstrate our approach. The simulation results show that this method, when based on linear, quadratic and nonparametric effect modification, can be more powerful than both a parametric multiplicative model fit and a fully nonparametric generalized additive model fit.

• 응용통계연구
• /
• 제29권1호
• /
• pp.267-277
• /
• 2016

본 연구에서는 잔차를 이용하여 오차항의 코플라 함수를 추정하는 문제를 고려하였다. 확률적 회귀모형을 개별모형으로 갖는 경우, 오차항 대신 잔차들의 경험적 분포함수를 이용하여 구한 코플라 모수에 대한 준모수적 추정량의 성질을 살펴보았으며, 이 추정량이 일치추정량이 되기 위한 조건을 구하였다. 응용사례로 코플라-자기회귀이동평균 모형을 다루었으며, 모의실험을 통해 자기회귀 근사를 통해 얻은 잔차를 이용하여 계산한 추정량의 성질도 살펴보았다.

• 응용통계연구
• /
• 제27권4호
• /
• pp.605-617
• /
• 2014

소지역추정은 작은 규모의 지역 또는 도메인에 작은 크기의 표본이 배정되어 추정의 정도가 좋지 않은 경우에 이를 극복하는 통계적 기법이다. 소지역추정에 흔히 사용되고 있는 모형기반 추정량은 MSE를 기초로 얻어지나 최근 상대오차를 이용한 소지역추정법도 연구되고 있다. 본 논문에서는 상대오차를 최소로 하는 소지역 추정량의 준모수적 접근법에 관하여 연구하였다. 즉 준모수혼합모형을 이용한 축소소지역추정량을 새롭게 제안하였다. 또한 Lee(1995)에서 제안된 모의실험 자료를 이용한 모의실험과 매월노동통계 자료를 이용한 사례연구를 통하여 기존의 추정량과 제안된 추정량의 우수성을 비교하였다.

• Communications for Statistical Applications and Methods
• /
• 제23권4호
• /
• pp.343-353
• /
• 2016

In doubly-censored data, an originating event time and a terminating event time are interval-censored. In certain analyses of such data, a researcher might be interested in the elapsed time between the originating and terminating events as well as regression modeling with risk factors. Therefore, in this study, we introduce a model evaluation method to measure the predictive ability of a model based on negative predictive values. We use a semiparametric estimate of the predictive accuracy to provide a simple and flexible method for model evaluation of doubly-censored survival outcomes. Additionally, we used simulation studies and tested data from a prostate cancer trial to illustrate the practical advantages of our approach. We believe that this method could be widely used to build prediction models or nomograms.

• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.809-817
• /
• 2012

We propose a semiparametric inference for a generalized varying coefficient partially linear model(VCPLM) for negative binomial data. The VCPLM is useful to model real data in that varying coefficients are a special type of interaction between explanatory variables and partially linear models fit both parametric and nonparametric terms. The negative binomial distribution often arise in modelling count data which usually are overdispersed. The varying coefficient function estimators and regression parameters in generalized VCPLM are obtained by formulating a penalized likelihood through smoothing splines for negative binomial data when the shape parameter is known. The performance of the proposed method is then evaluated by simulations.

• Journal of the Korean Statistical Society
• /
• 제32권1호
• /
• pp.85-92
• /
• 2003

We consider the partially linear model E(Y) : X$^{t}$ $\beta$＋η(Z) when the X's are measured with additive error. The semiparametric likelihood estimation ignoring the measurement error gives inconsistent estimator for both $\beta$ and η(.). In this paper we suggest the SIMEX estimator for f to correct the bias induced by measurement error, and explore its properties. We show that the rational linear extrapolant is proper in extrapolation step in the sense that the SIMEX method under this extrapolant gives consistent estimator It is also shown that the SIMEX estimator is asymptotically equivalent to the semiparametric version of the usual parametric correction for attenuation suggested by Liang et al. (1999) A simulation study is given to compare two variance estimating methods for SIMEX estimator.