• Journal of the Korean Statistical Society
• /
• 제26권3호
• /
• pp.299-308
• /
• 1997

In the simple linear regression model Y = .alpha.$_{0}$ + .beta.$_{0}$Z + .epsilon. under the right censorship of the response variables, the estimation of the mean lifetime E(Y) is an interesting problem. In this paper we propose a method of estimating E(Y) based on the observations modified by the arguments of Buckley and James (1979). It is shown that the proposed estimator is consistent and our proposed procedure in the simple linear regression case can be naturally extended to the multiple linear regression. Finally, we perform simulation studies to compare the proposed estimator with the estimator introduced by Gill (1983).83).

• Journal of the Korean Data and Information Science Society
• /
• 제25권5호
• /
• pp.1151-1160
• /
• 2014

In DNA microarray studies, the number of genes far exceeds the number of samples and the gene expression measures are highly correlated. Partial least squares regression (PLSR) is one of the popular methods for dimensional reduction and known to be useful for the classifications of microarray data by several studies. In this study, we suggest a modified version of the partial least squares regression to analyze gene expression data with survival information. The method is designed as a new gene selection method using PLSR with an iterative procedure of imputing censored survival time. Mean square error of prediction criterion is used to determine the dimension of the model. To visualize the data, plot for variables superimposed with samples are used. The method is applied to two microarray data sets, both containing survival time. The results show that the proposed method works well for interpreting gene expression microarray data.

• 응용통계연구
• /
• 제29권7호
• /
• pp.1231-1246
• /
• 2016

경쟁위험에 대한 연구 중 주로 쓰이는 방법은 Cause-specific 위험 모형과 subdistribution을 이용한 비례 위험 모형 방법이다. 그 이후에도 많은 모형이 제시되었지만, 추정 방법 면에서 설명력이 부족하거나 알고리즘으로 구현하기 어려운 단점을 가지고 있어서 잘 활용되고 있지 않다. 이 논문에서는 Cause-specific 위험 모형, subdistribution을 이용한 비례 위험 모형과 비교적 최근에 제시된 이항 회귀 모형(direct binomial model), 절대 위험 회귀 모형(absolute risk regression model), Eriksson 등 (2015)의 비례 오즈 모형(proportional odds model)을 소개하고 추정 방법을 간단히 설명하고자 한다. 각 모형에 대하여 SAS와 R을 이용한 활용 방법을 제시하고, 두 가지 경쟁위험이 존재하는 데이터를 R을 이용하여 분석하였다.

• 산업진흥연구
• /
• 제7권1호
• /
• pp.1-8
• /
• 2022

본 논문은 중도중단 데이터가 포함된 생존데이터의 경우 적용할 수 있는 기계학습 방법에 대해 살펴보았다. 우선 탐색적인 자료분석으로 각 특성에 대한 분포, 여러 특성들 간의 관계 및 중요도 순위를 파악할 수 있었다. 다음으로 독립변수에 해당하는 여러 특성들과 종속변수에 해당하는 특성(사망여부) 간의 관계를 분류문제로 보고 logistic regression, K nearest neighbor 등의 기계학습 방법들을 적용해본 결과 적은 수의 데이터이지만 통상적인 기계학습 결과에서와 같이 logistic regression보다는 random forest가 성능이 더 좋게 나왔다. 하지만 근래에 성능이 좋다고 하는 artificial neural network나 gradient boost와 같은 기계학습 방법은 성능이 월등히 좋게 나오지 않았는데, 그 이유는 주어진 데이터가 빅데이터가 아니기 때문인 것으로 판명된다. 마지막으로 Kaplan-Meier나 Cox의 비례위험모델과 같은 통상적인 생존분석 방법을 적용하여 어떤 독립변수가 종속변수 (t_i, δ_i)에 결정적인 영향을 미치는지 살펴볼 수 있었으며, 기계학습 방법에 속하는 random forest를 중도중단 데이터가 포함된 생존데이터에도 적용하여 성능을 평가할 수 있었다.

• 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.

• Communications for Statistical Applications and Methods
• /
• 제8권1호
• /
• pp.127-135
• /
• 2001

In this paper, we consider a simple method for testing the assumption of independent censoring on the basis of a Cox proportional hazards regression model with a time-dependent covariate. This method involves a two-stage sampling in which a random subset of censored observations is selected and followed-up until their true survival times are observed. Lee and Wolfe(1998) proposed an adjusted estimate of the survivor function for the dependent censoring under a proportional hazards alternative. This paper extends their result to obtain a bootstrap confidence interval for the adjusted survivor function under the dependent censoring. The proposed procedure is illustrated with an example of a clinical trial for lung cancer analysed in Lee and Wolfe(1998).

• Journal of Gastric Cancer
• /
• 제14권4호
• /
• pp.259-265
• /
• 2014

Purpose: Survival analysis of gastric cancer patients requires knowledge about factors that affect survival time. This paper attempted to analyze the survival of patients with incomplete registered data by using imputation methods. Materials and Methods: Three missing data imputation methods, including regression, expectation maximization algorithm, and multiple imputation (MI) using Monte Carlo Markov Chain methods, were applied to the data of cancer patients referred to the cancer institute at Imam Khomeini Hospital in Tehran in 2003 to 2008. The data included demographic variables, survival times, and censored variable of 471 patients with gastric cancer. After using imputation methods to account for missing covariate data, the data were analyzed using a Cox regression model and the results were compared. Results: The mean patient survival time after diagnosis was $49.1{\pm}4.4$ months. In the complete case analysis, which used information from 100 of the 471 patients, very wide and uninformative confidence intervals were obtained for the chemotherapy and surgery hazard ratios (HRs). However, after imputation, the maximum confidence interval widths for the chemotherapy and surgery HRs were 8.470 and 0.806, respectively. The minimum width corresponded with MI. Furthermore, the minimum Bayesian and Akaike information criteria values correlated with MI (-821.236 and -827.866, respectively). Conclusions: Missing value imputation increased the estimate precision and accuracy. In addition, MI yielded better results when compared with the expectation maximization algorithm and regression simple imputation methods.

• 한국도로학회논문집
• /
• 제14권6호
• /
• pp.139-148
• /
• 2012

PURPOSES : The accuracy of travel time information is a key measure of effectiveness and reliability of advanced traveler information systems. This study aims at investigating drivers' perception on the acceptable level of information accuracy and their corresponding valuations. METHODS : A questionnaire survey was executed for collecting driver perception data to capture the expectation level of travel time information provided and their willingness to pay for the information. A Tobit model was adopted for exploring the relationship among the acceptable level, driver socioeconomic characteristics and travel attributes. Since drivers' willingness to pay for accurate travel time information can be different according to their travel lengths, a piecewise linear regression model was developed to capture the sensitivity of values of travel time information to travel length. RESULTS : The analysis results suggest that trip purpose and travel time are two dominant factors to determine drivers' acceptable level of travel time information. For business and short trips, drivers want more accurate information than for non-business and long trips. Drivers' willingness to pay for travel time information also varies depending on their incomes, trip purposes and travel lengths. The results also show that drivers' valuation of travel time information provided is sensitive to their travel length. For longer trips, drivers are less sensitive to travel time information and then put less value on the information provided. CONCLUSIONS : Censored nonlinear regression models are developed to estimate drivers' acceptable accuracy for travel time information and their valuation using questionnaire survey data. The findings on drivers perception to the required accuracy of travel time information and their corresponding willingness to pay can be used in the design and deployment of advanced traveler information system to improve its effectiveness and usefulness through high compliance.

• Management Science and Financial Engineering
• /
• 제11권3호
• /
• pp.125-136
• /
• 2005

As the importance of time-based competition is increasing, information systems for supporting the immediate decision making is strongly required. Especially high -tech product firms are under extreme pressure of rapid response to the demand side due to relatively short life cycle of the product. Therefore, the objective of our research is proposing a framework of estimating demand curve based on e-auction data, which is extremely easy to access and well reflect the limited demand curve in that channel. Firstly, we identify the advantages of using e-auction data for full demand curve estimation and then verify it using Agent-Eased-Modeling and Tobin's censored regression model.

• 산경연구논집
• /
• 제12권8호
• /
• pp.33-39
• /
• 2021

Purpose: The influence of age on time preference is not identified in the usual cross-sectional analysis. This study aims to test whether age affects time preference after controlling for the effects of individual heterogeneity including cohort effects. Research design, data and methodology: Drawing on a nationally representative panel dataset of Indonesians, we estimate the effects of age on time preference after controlling for unobserved individual heterogeneity as well as potential cohort effects. We measure time preference exploiting information on two sets of multiple price lists: one for a one-year delay, and the other for a five-year delay. Results: When we controlled for time-invariant individual characteristics, including birth cohort effects in a fixed effects model, the older men and women were more patient in a linear fashion, particularly when the delay was longer. To highlight the importance of controlling for individual fixed effects, we repeated the specification without controlling for individual fixed effects in OLS or censored maximum likelihood regression; we found no relation between age and impatience in men or women and for a one or five-year delay. Conclusions: The older men and women are more patient, and time preferences are correlated with unobserved individual heterogeneity.