• The Korean Journal of Applied Statistics
• /
• v.21 no.4
• /
• pp.603-613
• /
• 2008

We often encounter the situation that discrete count data have a large portion of zeros. In this case, it is not appropriate to analyze the data based on standard regression models such as the poisson or negative binomial regression models. In this article, we consider Bayesian analysis for two commonly used models. They are zero-inflated poisson and negative binomial regression models. We use the Bayes factor as a model selection tool and computation is proceeded via Markov chain Monte Carlo methods. Crash count data are analyzed to support theoretical results.

• The Korean Journal of Applied Statistics
• /
• v.15 no.1
• /
• pp.57-71
• /
• 2002

We introduce three types of exponential survival models, such as simple model, change-point model and finite mixture model in this paper. Among these models, in order to choose the best model, the model choice method is proposed using Gelfand and Ghosh(1998)'s idea. Then to avoid the computational difficulties, data augmentation method (Tanner and Wong, 1987) and Gibbs sampler (Gelfand and Smith, 1990) are employed. Our methodology is applied to both simulated data and Stangl (1991)'s On-impramint Hydrochloride data.

• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.135-140
• /
• 2000

베이지안 신경망 모형(Bayesian Neural Networks Models)에서 주어진 입력값(input)은 블랙 박스(Black-Box)와 같은 신경망 구조의 각 층(layer)을 거쳐서 출력값(output)으로 계산된다. 새로운 입력 데이터에 대한 예측값은 사후분포(posterior distribution)의 기대값(mean)에 의해 계산된다. 주어진 사전분포(prior distribution)와 학습데이터에 의한 가능도함수(likelihood functions)를 통해 계산되어진 사후분포는 매우 복잡한 구조를 갖게 됨으로서 기대값의 적분계산에 대한 어려움이 발생한다. 이때 확률적 추정에 의한 근사 방법인 몬테칼로 적분을 이용한다. 이러한 방법으로서 Hybrid Monte Carlo 알고리즘은 우수한 결과를 제공하여준다(Neal 1996). 본 논문에서는 Hybrid Monte Carlo 알고리즘과 기존에 많이 사용되고 있는 Gibbs sampling, Metropolis algorithm, 그리고 Slice Sampling등의 몬테칼로 방법들을 비교한다.

• The Korean Journal of Applied Statistics
• /
• v.7 no.1
• /
• pp.1-17
• /
• 1994

This article basically considers the calibration problem with censored data from the Bayesian point of view. The Gibbs sampling method is discussed to solve the difficulty encountered in computing the posterior distribution. Also presented is an approach for impementing the Gibbs sampling in actual data situation with the estimation procedures.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.833-845
• /
• 2011

This paper conducts a statistical analysis of extreme values for both daily log-returns and daily negative log-returns, which are computed using a collection of KOSPI data from January 3, 1998 to August 31, 2011. The Poisson-GPD model is used as a statistical analysis model for extreme values and the maximum likelihood method is applied for the estimation of parameters and extreme quantiles. To the Poisson-GPD model is also added the Bayesian method that assumes the usual noninformative prior distribution for the parameters, where the Markov chain Monte Carlo method is applied for the estimation of parameters and extreme quantiles. According to this analysis, both the maximum likelihood method and the Bayesian method form the same conclusion that the distribution of the log-returns has a shorter right tail than the normal distribution, but that the distribution of the negative log-returns has a heavier right tail than the normal distribution. An advantage of using the Bayesian method in extreme value analysis is that there is nothing to worry about the classical asymptotic properties of the maximum likelihood estimators even when the regularity conditions are not satisfied, and that in prediction it is effective to reflect the uncertainties from both the parameters and a future observation.

• The Korean Journal of Applied Statistics
• /
• v.18 no.1
• /
• pp.67-81
• /
• 2005

Most of the previously proposed methods for the frailty model do not work well when there are many tied observations. This is partly because the empirical likelihood used is not suitable for tied observations. In this paper, we propose a new method for the frailty model with many ties. The proposed method obtains the posterior distribution of the parameters using the binomial form empirical likelihood and Bayesian bootstrap. The proposed method yields stable results and is computationally fast. To compare the proposed method with the maximum marginal likelihood approach, we do simulations.

• The Korean Journal of Applied Statistics
• /
• v.31 no.6
• /
• pp.677-692
• /
• 2018

The total fertility rate of Korea was 1.05 in 2017, showing a return to the 1.08 level in the year 2005. 1.05 is a very low fertility level that is far from replacement level fertility or safety zone 1.5. The number may indicate a low fertility trap. It is therefore important to predict fertility than at any other time. In the meantime, we have predicted the age-specific fertility rate and total fertility rate by various statistical methods. When the data trend is disconnected or fluctuating, it applied a nonparametric method applying the smoothness and weight. In addition, the Bayesian method of using the pre-distribution of fertility rates in advanced countries with reference to the three-stage transition phenomenon have been applied. This paper examines which method is reasonable in terms of precision and feasibility by applying estimation, forecasting, and comparing the results of the recent variability of the Korean fertility rate with parametric, non-parametric and Bayesian methods. The results of the analysis showed that the total fertility rate was in the order of KOSTAT's total fertility rate, Bayesian, parametric and non-parametric method outcomes. Given the level of TFR 1.05 in 2017, the predicted total fertility rate derived from the parametric and nonparametric models is most reasonable. In addition, if a fertility rate data is highly complete and a quality is good, the parametric model approach is superior to other methods in terms of parameter estimation, calculation efficiency and goodness-of-fit.

• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.365-377
• /
• 2020

We discuss latent factor regression when constructing a common structure inherent among explanatory variables to solve multicollinearity and use them as regressors to construct a linear model of a response variable. Bayesian estimation with LASSO prior of a large penalty parameter to construct a significant factor loading matrix of intrinsic interests among infinite latent structures. The estimated factor loading matrix with estimated other parameters can be inversely transformed into linear parameters of each explanatory variable and used as prediction models for new observations. We apply the proposed method to Product Service Management data of HBAT and observe that the proposed method constructs the same factors of general common factor analysis for the fixed number of factors. The calculated MSE of predicted values of Bayesian latent factor regression model is also smaller than the common factor regression model.

• Convergence Security Journal
• /
• v.6 no.3
• /
• pp.117-126
• /
• 2006

Bayesian inference and model selection method for software reliability growth models are studied. Software reliability growth models are used in testing stages of software development to model the error content and time intervals between software failures. In this paper, could avoid multiple integration using Gibbs sampling, which is a kind of Markov Chain Monte Carlo method to compute the posterior distribution. Bayesian inference for general order statistics models in software reliability with diffuse prior information and model selection method are studied. For model determination and selection, explored goodness of fit (the error sum of squares), trend tests. The methodology developed in this paper is exemplified with a software reliability random data set introduced by of Weibull distribution(shape 2 & scale 5) of Minitab (version 14) statistical package.

• The Korean Journal of Applied Statistics
• /
• v.13 no.1
• /
• pp.115-131
• /
• 2000

이 논문에서는 평균-이동모형(mean-shift model)을 이상점을 위한 대립모형으로 사용하여 변량모형(random effect model)에서의 이상점 검출을 위한 베이즈인자(Bayes factor)를 제시한다. 그러나 가능한 사전 정보가 없어서 무정보사전분포(noninformative prior distribution)가 사용되어야만 할 때, 대부분의 무정보사전분포는 부적절분포(improper distribution)이기 때문에 베이즌 인자에는 사전분포로부터 나온 미지의 상수가 포함되어 잇다. 이 문제를 해결하기 위해 이 논문에서는 Berger와 Pericchi (1996)가 제시한 내재베이즈인자(the intrinsic Bayes factor;IBF)를 사용한다. 또한 이 베이즈인자를 계산상 어려움을 해결하기 위해 Verdinellidh Wasserman(1995)의 일반화 세비디지키 밀도비를 이용하여 수정하고 이것을 이용하여 이상점을 검출하는 방법을 제시한다. 마지막으로 인위적으로 이상점을 포함하고 있는 데이터를 만들고 제시된 방법으로 가상실험을 하고 또한 실제 데이터에서 제시한 방법으로 이상점을 찾아보았다.