• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.83-97
• /
• 2019

As society develops, the dissemination of microdata has increased to respond to diverse analytical needs of users. Analysis of microdata for policy making, academic purposes, etc. is highly desirable in terms of value creation. However, the provision of microdata, whose usefulness is guaranteed, has a risk of exposure of personal information. Several methods have been considered to ensure the protection of personal information while ensuring the usefulness of the data. One of these methods has been studied to generate and utilize synthetic data. This paper aims to understand the synthetic data by exploring methodologies and precautions related to synthetic data. To this end, we first explain muptiple imputation, Bayesian predictive model, and Bayesian bootstrap, which are basic foundations for synthetic data. And then, we link these concepts to the construction of fully/partially synthetic data. To understand the creation of synthetic data, we review a real longitudinal synthetic data example which is based on sequential regression multivariate imputation.

• 한국경영정보학회:학술대회논문집
• /
• 2007.06a
• /
• pp.318-323
• /
• 2007

국방부에서 발표한 ‘국방개혁에 관한 법률’ 에 따라 2014년까지 현역병들에 대한 복무기간이 단계적으로 단축될 예정이다. 이에 따라 좀 더 효율적인 직무교육 방안이 필요하게 되어, ‘차등제 교육’을 시행하고 있다. 이 교육의 효과를 향상시키기 위해서는 훈련병들의 예상 학업 성취도를 미리 정확하게 예측하는 것이 필수적이다. 따라서, 본 연구에서는 입교 초기에 얻을 수 있는 신병들의 제한된 자료들을 이용하여 교육 성취도 예측 모형을 개발하였다. 본 모형의 목적 변수는 ‘일반관리 인원’, ‘집중관리 인원’의 값을 갖는 이진형 성취집단 변수이며, 사용된 기법은 k-means 군집기볍과 Decision Tree 기법을 혼합한 모형, k-means 군집기법과 Neural Network 기법을 혼합한 모형, Decision Tree 모형, Neural Network 모형, Bayesian 모형, 그리고 Logistic 모형 등을 사용하였다. 그 결과 k-means 군집기법과 Decision Tree를 혼합한 모형이 가장 좋은 예측력올 보이는 것으로 나타났다. 이러한 교육 성취집단 예측 모형은 향후 군에서 이루어지는 다양한 교육 프로그램에 적극적으로 이용될 수 있을 것으로 기대된다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.32 no.2
• /
• pp.141-152
• /
• 2006

This paper presents the development of a decision model to examine the optimal repair action for a deteriorating system. In order to make a reasonable decision, it is necessary to perform an analysis of the uncertainties embedded in deterioration and to evaluate the repair actions based on the expected future cost. Focusing on the power law failure model, the uncertainties related to deterioration are analyzed based on the Bayesian approach. In addition, we develop a decision model for the optimal repair action by applying a repair cost function. A case study is given to illustrate a decision-making process by analyzing the loss incurred due to deterioration.

• Environmental and Resource Economics Review
• /
• v.11 no.1
• /
• pp.99-119
• /
• 2002

As an alternative to classical maximum likelihood approach for analyzing dichotomous choice contingent valuation (DCCV) data, this paper develops a Bayesian approach. By using the idea of Gibbs sampling and data augmentation, the approach enables one to perform exact inference for DCCV models. A by-product from the approach is welfare measure, such as the mean willingness to pay, and its confidence interval, which can be used for policy analysis. The efficacy of the approach relative to the classical approach is discussed in the context of empirical DCCV studies. It is concluded that there appears to be considerable scope for the use of the Bayesian analysis in dealing with DCCV data.

• Journal of Korean Institute of Industrial Engineers
• /
• v.24 no.4
• /
• pp.485-492
• /
• 1998

Most forecasting models often fail to produce appropriate forecasts because we build a model based on the assumption of the data being generated from the only one stochastic process. However, in many real problems, the time series data are generated from one stochastic process for a while and then abruptly undergo certain structural changes. In this paper, we assume the basic underlying process is the simple state-space model with random level and deterministic drift but interrupted by three types of exogenous shocks: level shift, drift change, outlier. A Bayesian procedure to detect, estimate and adapt to the structural changes is developed and compared with simple, double and adaptive exponential smoothing using simulated data and the U.S. leading composite index.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.309-309
• /
• 2019

기후예측모델을 통해 일단위 강수의 예측정보가 제공되고 있지만, 실제 강수량자료와 시공간적 편의로 인해 수문학적 활용은 한계가 있다. 일반적으로 기후모델의 시공간적 해석 규모 및 예측정확성을 고려할 때 계절단위에서 예측정보의 활용이 가장 현실적인 것으로 알려지고 있다. 그러나 수문해석 시 시공간적 해상도가 낮아 직접적인 활용은 어려운 상황이며, 수문해석 모형의 입력자료로 활용 시 편의보정 및 상세화 과정이 일반적으로 요구된다. 본 연구에서는 기후모델로부터 얻은 강우예측결과에 Bayesian 모델 기반의 편의보정-상세화 기법을 개발하여 강우예측정보의 활용성을 개선하고자 한다. 이 과정에서 Bayesian Copula 모델을 이용한 이변량 형태의 예측강수의 검보정 방법을 개발하였으며, 특히 기후모델 이외의 기상 상태변량인 해수면온도(sea surface temperature, SST)를 예측인자로 추가하여 Hybrid 형태의 계절 앙상블 강우예측모델을 개발하고자 한다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.1
• /
• pp.29-41
• /
• 2017

Design rainfalls are one of the most important hydrologic data for river management, hydraulic structure design and risk analysis. The design rainfalls are first estimated by a point frequency analysis and the IDF (intensity-duration-frequency) curve is then constructed by a nonlinear regression to either interpolate or extrapolate the design rainfalls for other durations which are not used in the frequency analysis. It has been widely recognised that the more reliable approaches are required to better account for uncertainties associated with the model parameters under circumstances where limited hydrologic data are available for the watershed of interest. For these reasons, this study developed a hierarchical Bayesian based GLM (generalized linear model) for a regional frequency analysis in conjunction with a scaling function of the parameters in probability distribution. The proposed model provided a reliable estimation of a set of parameters for each individual station, as well as offered a regional estimate of the parameters, which allow us to have a regional IDF curve. Overall, we expected the proposed model can be used for different aspects of water resources planning at various stages and in addition for the ungaged basin.

• The Korean Journal of Applied Statistics
• /
• v.33 no.2
• /
• pp.203-213
• /
• 2020

The self-controlled case series (SCCS) study measures the relative risk of exposure to exposure period by setting the non-exposure period of the patient as the control period without a separate control group. This method minimizes the bias that occurs when selecting a control group and is often used to measure the risk of adverse events after taking a drug. This study used SCCS to examine the increased risk of side effects when two or more drugs are used in combination. A conditional Poisson model is assumed and analyzed for drug interaction between the narcotic analgesic, tramadol and multi-frequency combination drugs. Bayesian inference is used to solve the overfitting problem of MLE and the normal or Laplace prior distributions are used to measure the sensitivity of the prior distribution.

• Journal of Korea Water Resources Association
• /
• v.53 no.2
• /
• pp.121-129
• /
• 2020

The increase of impervious surface and development along the river due to urbanization not only causes an increase in the number of associated flood risk factors but also exacerbates flood damage, leading to difficulties in flood management. Flood control measures should be prioritized based on various geographical information in urban areas. In this study, a probabilistic flood hazard assessment was applied to flood-prone areas near an urban river. Flood hazard maps were alternatively considered and used to describe the expected inundation areas for a given set of predictors such as elevation, slope, runoff curve number, and distance to river. This study proposes a Bayesian logistic regression-based flood risk model that aims to provide a probabilistic risk metric such as population-at-risk (PAR). Finally, the logistic regression model demonstrates the probabilistic flood hazard maps for the entire area.

• Environmental and Resource Economics Review
• /
• v.19 no.2
• /
• pp.215-242
• /
• 2010

This study constructs a bayesian neoclassical DSGE model that applies oil usage. The model includes technology shocks, oil price shocks, and shocks to energy policies as exogenous driving forces. First, this study aims to analyze the roles of these exogenous shocks in the Korean business cycle. Second, this study examines the effects of long-term changes in the energy consumption structure, including the reduction in oil use as a share of energy consumption and improvement in oil efficiency. In the case of oil price shocks, results show that these shocks exert recessionary pressure on the economy in line with those obtained in the previous literature. On the other hand, shocks to energy policies, which reduce oil consumption per capital, result in opposite consequences to oil price shocks, decreasing oil consumption. Also, counterfactual exercises show that long-term changes in the energy consumption structure would mitigate the contractionary effects of oil price shocks.