• 응용통계연구
• /
• 제11권1호
• /
• pp.163-175
• /
• 1998

위험률 변화점모형에서 특별한 함수형이나 분포함수에 대한 가정을 하지 않는 일반적인 모형을 고려하였다. 이러한 모형은 지금까지 주로 다루어 왔던 상수항 위험률의 변화점모형뿐만 아니라 여러 유형의 변화점모형을 내포한다. 중도절단된 자료하에서 위험률 변화점에 관한 모수적 모형을 가정하지 않고 변화점 이전과 이후의 넬슨(Nelson) 누적위험함수 추정량의 기울기 차를 이용하여 추정량을 제안하고, 그의 점근적 성질을 규명한다. 붓스트랩 추정량의 일치성과 점근분포를 유도하고, 몇가지 분포함수의 경우에 몬테칼로 모의실험을 통해 제안된 방법의 경험적 성질을 살펴보았다. 또한, 심장병 이석환자의 생존시간 자료를 통해 변화점을 추정하고 추정량의 붓스트랩 분포를 구하였다.

• 한국디지털정책학회:학술대회논문집
• /
• 한국디지털정책학회 2006년도 추계학술대회
• /
• pp.327-336
• /
• 2006

The odds ratio is used for assessing the disease-exposure association, because epidemiological data for case-control of cohort studies are often summarized into 2 ${\times}$ 2 tables. In this paper we define the odds ratio function(ORF) that extends odds ratio used on discrete survival event data to continuous survival time data and propose estimation procedures with censored data. The first one is a nonparametric estimator based on the Nelson-Aalen estimator of comulative hazard function, and the others are obtained using the concept of empirical odds ratio. Asymptotic properties such as consistency and weak convergence results are also provided. The ORF provides a simple interpretation and is comparable to survival function or comulative hazard function in comparing two groups. The mean square errors are investigated via Monte Carlo simulation. The result are finally illustrated using the Melanoma data.

• Communications for Statistical Applications and Methods
• /
• 제26권6호
• /
• pp.635-644
• /
• 2019

Class probability is a fundamental target in classification that contains complete classification information. In this article, we propose a class probability estimation method when the predictor is functional. Motivated by Wang et al. (Biometrika, 95, 149-167, 2007), our estimator is obtained by training a sequence of functional weighted support vector machines (FWSVM) with different weights, which can be justified by the Fisher consistency of the hinge loss. The proposed method can be extended to multiclass classification via pairwise coupling proposed by Wu et al. (Journal of Machine Learning Research, 5, 975-1005, 2004). The use of FWSVM makes our method model-free as well as computationally efficient due to the piecewise linearity of the FWSVM solutions as functions of the weight. Numerical investigation to both synthetic and real data show the advantageous performance of the proposed method.

• Communications for Statistical Applications and Methods
• /
• 제8권1호
• /
• pp.229-235
• /
• 2001

Let {X$\_$n/, n$\geq$1] be a stationary sequence of negatively associated random variables with distribution function F(x)=P(X$_1$$\leq$x). The empirical distribution function F$\_$n/(x) based on X$_1$, X$_2$,....., X$\_$n/ is proposed as an estimator for F$\_$n/(x). Strong consistency and asymptotic normality of F$\_$n/(x) are studied. We also apply these ideas to estimation of the survival function.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제31권2호
• /
• pp.133-140
• /
• 1992

The problem of estimating a smooth distribution function F at a point $\tau$ based on randomly right censored data is treated under certain smoothness conditions on F . The asymptotic performance of a certain class of kernel estimators is compared to that of the Kap lan-Meier estimator of F($\tau$). It is shown that the .elative deficiency of the Kaplan-Meier estimate. of F($\tau$) with respect to the appropriately chosen kernel type estimate. tends to infinity as the sample size n increases to infinity. Strong uniform consistency and the weak convergence of the normalized process are also proved.

• Journal of the Korean Data and Information Science Society
• /
• 제18권3호
• /
• pp.669-678
• /
• 2007

An algorithm to detect the number of discontinuity points of the variance function in regression model is proposed. The proposed algorithm is based on the left and right one-sided kernel estimators of the second moment function and test statistics of the existence of a discontinuity point coming from the asymptotic distribution of the estimated jump size. The finite sample performance is illustrated by simulated example.

• Journal of the Korean Statistical Society
• /
• 제28권3호
• /
• pp.359-370
• /
• 1999

We considered a change-point hazard rate model generalizing constant hazard rate model. This type of model is very popular in the sense that the Weibull and exponential distributions formulating survival time data are the special cases of it. Maximum likelihood estimation and the asymptotic properties such as the consistency and its limiting distribution of the change-point estimator were discussed. A parametric bootstrap method for finding confidence intervals of the unknown change-point was also suggested and the proposed method is explained through a practical example.

• Communications for Statistical Applications and Methods
• /
• 제6권2호
• /
• pp.467-478
• /
• 1999

This paper deals with the problem of change-point estimation where there is one level change in location with iid errors. A change-point estimator using rank average is proposed with the proof of its consistency. A comparison study of various change-point estimators is done by simulation on the mean the proportion and the variance when the errors are from the normal and the double exponential distributions.

• Journal of the Korean Data and Information Science Society
• /
• 제6권2호
• /
• pp.65-76
• /
• 1995

The importance of left truncated and right censoring cases has considered for better information in medical follow-up and engineering life testing studies. We propose some estimation procedure for the mean residual life function with consistency and asymptotic normality on the left truncated and right censoring model. And then, the comparision with Kaplan-Meier estimator ignoring the left truncated effect and the small sample properities are investigated by asymptotic biases and M.S.E.'s thresh Monte Carlo study.

• Journal of the Korean Data and Information Science Society
• /
• 제10권1호
• /
• pp.207-214
• /
• 1999

Optimal estimating functions for a class of autoregressive models with GARCH(1,1) error are discussed. The asymptotic properties of the estimator as the solution of the optimal estimating equation are investigated for the models. We have also some simulation results which suggest that the proposed optimal estimators have smaller sample variances than those of the Conditional least-squares estimators under the heavy-tailed error distributions.