• Proceedings of the Korean Statistical Society Conference
• /
• 2002.11a
• /
• pp.69-73
• /
• 2002

본 발표에서는 베이지안이 생각하는 확률의 개념을 상호교환성(exchangeability)의 가정아래 어떻게 확장되어 해석되는지를 소개하고, 빈도학자들의 접근방법과 비교함으로서 베이지안에서 생각하는 확률이 어떠한 특징을 가지고 있는지를 설명하고자 하였다. 또한 Efron에 의하여 지적된 베이지안의 네 가지 문제점에 대하여 논의하고 특별히 과학적 객관성(scientific objectivism)의 한계점과 이러한 한계점을 베이지안에서 어떻게 해결하고 있는지에 대하여 논의하였다. 일반적으로 과학적 객관성에 대한 한계점은 빈도학자들의 방법론에서도 존재하게 된다. 즉, 연구자가 가설을 설정하고 이에 맞는 실험설계를 하고 유의수준을 설정하고 p값을 이용하여 의사결정을 내리는 모든 단계에서 연구자의 주관성이 들어갈 수밖에 없게 된다는 것이다. 베이지안 방법론에서는 이러한 비객관적인 체계를 인정하고 파악하여 사전확률(prior)에 포함시킴으로서 이를 객관적인 자료인 가능도함수(likelihood function)와 혼합하여 추론이나 의사결정을 진행하게 된다. 마지막으로 베이지안 학자들의 최근 객관적인 사전확률에 대한 다양한 형태의 연구를 소개하는 것으로 발표를 마무리하고자 한다.

• Journal of the Korean Society for information Management
• /
• v.8 no.2
• /
• pp.74-95
• /
• 1991

The purpose of this paper is to study to the nature and function of scientific inquiry and research methods, particularly as they apply to the influencing factors in library schools. For the variables used, they are the average cost per student for training, the total percentage placement in library and library-related work, the average salary for year, and the type of schools. In interpreting data using, t -test analysis of correlation and regression are applied.

• The Korean Journal of Applied Statistics
• /
• v.2 no.2
• /
• pp.61-78
• /
• 1989

We consider the step stress life testing which has been developed in order to perform the life testing of the units whose normal life time is long within a reasonable amount of time. The models suggested for statistical analysis of the data obtained form the stress life testing are reviewed and a model which contains these models in some respect is suggested. The statistical inference based on the suggested model is done using maximum likelihood and weighted least square estimates. Finally we review the design of the simple step stress life testing and extend the result to the censoring case.

• The Korean Journal of Applied Statistics
• /
• v.2 no.2
• /
• pp.27-35
• /
• 1989

This article deals with statistical inference on Interleukin-2 titer of which the clinical applications to cancer immunotherapy and some immunodeficiency diseases have been widely tried. A Linear model and the Bayesian approach are used to explain the bioassay which performs the measurements of IL-2 activity from an patient and an inference procedure including confidence intervals for the IL-2 titer of the patient through comparision with the Standard IL-2 is suggested and a real case of example is illustrated.

• The Korean Journal of Applied Statistics
• /
• v.13 no.2
• /
• pp.237-245
• /
• 2000

In this paper we consider a logistic regression model based on two-stage case-control sampling and study the Weighted Exogeneous Sampling Maximum Likelihood(WESML) method to get an asymptotically normal estimates of the parameters in a logistic regression model. A numerical example is carried out to demonstrate the differences between the Conditional Maximum Likelihood(CML) estimates and the WESML estimates for two-stage case-control data.

• The Korean Journal of Applied Statistics
• /
• v.17 no.1
• /
• pp.145-151
• /
• 2004

PROC MIXED fits a variety of mixed models to data and enables one to use these fitted models to make statistical inferences about the data. However, the simulation study in this article shows that PROC MIXED using REML estimators provides one with a confidence interval, that does not keep the stated confidence coefficients, on sums of two variance components in the simple regression model with unbalanced nested error structure which is a mixed model.

• The Korean Journal of Applied Statistics
• /
• v.7 no.2
• /
• pp.9-19
• /
• 1994

Beta-binomial model, which is reparametrized in terms of the mean probability $\mu$ of a positive deagnosis and the $\kappa$ of agreement, is widely used in psychology. When $\mu$ is close to 0, inference about $\kappa$ become difficult because likelihood function becomes constant. We consider Bayesian approach in this case. To apply Bayesian analysis, Gibbs sampler is used to overcome difficulties in integration. Marginal posterior density functions are estimated and Bayesian estimates are derived by using Gibbs sampler and compare the results with the one obtained by using numerical integration.

• Communications of Mathematical Education
• /
• v.11
• /
• pp.107-126
• /
• 2001

우리는 흔히 21C를 정보화 시대라고 하며 우리에게 주어지는 정보들 또한 일기예보와 같은 일상적인 분야에서 여론 조사와 같은 전문적인 분야에 이르기까지 아주 다양하다. 이런 정보들은 통계영역과 아주 밀접하며 이런 정보들을 통계적으로 바르게 해석하고 추론하여 일반화하는 등 일련의 과정들을 요구한다. 이런 상황아래 본 연구에서는 6차 초등학교 수학 교과서에서 여러 통계학 영역 중 그래프 형태로 가장 먼저 도입되는 막대그래프에 중점을 두어 현행 교과서에서 학습 내용과 학습 과정의 문제점에는 어떤 것이 있으며 아울러 그래프 이해력에 필요한 요소나 인지적 사고 능력, 그래프 이해력의 수준을 알아보고, 이를 바탕으로 여러 문헌을 통해 본 연구자가 나름대로 구안한 통계적 기법을 사용한 교수 ${\cdot}$ 학습 과정을 실험반에 적용한 후 그래프 이해력 사전 ${\cdot}$ 사후 검사를 비교함으로써 통계적 기법을 사용한 교수 ${\cdot}$ 학습 과정이 그래프 이해력에 어떠한 영향을 미치는지 알아보고자 한다.

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.119-133
• /
• 1999

A Markov Chain Monte Carlo method is developed to compute the software reliability model. We consider computation problem for determining of posterior distibution in Bayseian inference. Metropolis algorithms along with Gibbs sampling are proposed to preform the Bayesian inference of the Mixed model with record value statistics. For model determiniation, we explored the prequential conditional predictive ordinate criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions. A numerical example with simulated data set is given.

• The Korean Journal of Applied Statistics
• /
• v.6 no.2
• /
• pp.435-442
• /
• 1993

For the generalized gamma distribution, exact inferences about quantiles need many computations involving complicated numerical integrations. This paper suggests approximate confidence intervals which are easily obtained by considering the alternative location-scale model. Also, these intervals are very accurate even for small sample size. Approximate confidence intervals about quantiles in the lognormal distribution are also considered. With type 2 censoring data, approximate confidence intervals can also be obtained directly by the suggested methods.