• Journal of the Korean Data and Information Science Society
• /
• 제26권6호
• /
• pp.1547-1555
• /
• 2015

We study the problem of nonignorable nonresponse in a two-way contingency table and there may be one or two missing categories. We describe a nonignorable nonresponse model for the analysis of two-way categorical table. One approach to analyze these data is to construct several tables (one complete and the others incomplete). There are nonidentifiable parameters in incomplete tables. We describe a hierarchical Bayesian model to analyze two-way categorical data. We use a nonignorable nonresponse model with Bayesian uncertainty analysis by placing priors in nonidentifiable parameters instead of a sensitivity analysis for nonidentifiable parameters. To reduce the effects of nonidentifiable parameters, we project the parameters to a lower dimensional space and we allow the reduced set of parameters to share a common distribution. We use the griddy Gibbs sampler to fit our models and compute DIC and BPP for model diagnostics. We illustrate our method using data from NHANES III data to obtain the finite population proportions.

• 한국수자원학회논문집
• /
• 제56권8호
• /
• pp.509-520
• /
• 2023

가뭄은 강수량, 토양수분 그리고 유출량 등 여러 가지 요인으로부터 영향을 받으며, 가뭄의 상황 판단을 위하여 다양한 가뭄지수가 널리 활용되고 있다. 가뭄지수의 산정에 활용되는 수문기상학적 자료와 가뭄지수 산정공식에 따라서 지수값은 달라지며, 가뭄 상황에 대한 판단에도 차이가 발생 할 수 있다. 본 연구에서는 국내외에서 널리 활용되는 표준강수지수(SPI)의 산정과정에서 결정해야 하는 강수량의 자료길이, 누적기간, 확률분포 모형, 매개변수 추정기법 등을 불확실성 영향 요인으로 가정하고, 각각의 조합에 대한 불확실성을 평균제곱근오차와 선형혼합모형(LMM)을 활용하여 평가하였다. 평균제곱오차는 SPI 산정과정에 발생되는 전반적인 오차를 추정하며, LMM은 영향 요인들의 상대적인 불확설성을 평가하는데 활용되었다. 그 결과, SPI 산정에 활용된 자료의 기간과 누적기간이 길어질수록 평균제곱오차가 감소하였다. LMM을 통하여 불확실성 영향요인들의 기여도를 비교한 결과, SPI의 불확실성에는 자료기간의 영향이 가장 크게 나타났다. 또한, 자료기간이 증가하면, 자료기간에 의한 불확실성은 감소하고 누적기간과 매개변수 추정기법에 의한 불확실성이 상대적으로 증가하였다. 본 연구 결과, SPI 산정과정에서 발생되는 불확실성을 줄이기 위해서는 장기간의 자료 확보가 우선이며, 자료의 특성을 적절히 반영하는 확률분포모형과 매개변수 추정기법이 적용되어야 한다.

• Nuclear Engineering and Technology
• /
• 제52권2호
• /
• pp.287-295
• /
• 2020

Group method of data handling (GMDH) is considered one of the earliest deep learning methods. Deep learning gained additional interest in today's applications due to its capability to handle complex and high dimensional problems. In this study, multi-layer GMDH networks are used to perform uncertainty quantification (UQ) and sensitivity analysis (SA) of nuclear reactor simulations. GMDH is utilized as a surrogate/metamodel to replace high fidelity computer models with cheap-to-evaluate surrogate models, which facilitate UQ and SA tasks (e.g. variance decomposition, uncertainty propagation, etc.). GMDH performance is validated through two UQ applications in reactor simulations: (1) low dimensional input space (two-phase flow in a reactor channel), and (2) high dimensional space (8-group homogenized cross-sections). In both applications, GMDH networks show very good performance with small mean absolute and squared errors as well as high accuracy in capturing the target variance. GMDH is utilized afterward to perform UQ tasks such as variance decomposition through Sobol indices, and GMDH-based uncertainty propagation with large number of samples. GMDH performance is also compared to other surrogates including Gaussian processes and polynomial chaos expansions. The comparison shows that GMDH has competitive performance with the other methods for the low dimensional problem, and reliable performance for the high dimensional problem.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2018년도 학술발표회
• /
• pp.151-151
• /
• 2018

The present study is aimed to quantifying the uncertainty in the general circulation model (GCM) selection and its impacts on hydrology studies in the basins. For this reason, 13 GCMs was selected among the 26 GCM models of the Fifth Assessment Report (AR5) scenarios. Then, the climate data and hydrologic data with two Representative Concentration Pathways (RCPs) of the best model (INMCM4) and worst model (HadGEM2-AO) were compared to understand the uncertainty associated with GCM models. In order to project the runoff, the Precipitation-Runoff Modelling System (PRMS) was driven to simulate daily river discharge by using daily precipitation, maximum and minimum temperature as inputs of this model. For simulating the discharge, the model has been calibrated and validated for daily data. Root mean square error (RMSE) and Nash-Sutcliffe Efficiency (NSE) were applied as evaluation criteria. Then parameters of the model were applied for the periods 2011-2040, and 2070-2099 to project the future discharge the five large basins of South Korea. Then, uncertainty caused by projected temperature, precipitation and runoff changes were compared in seasonal and annual time scale for two future periods and RCPs compared to the reference period (1976-2005). The findings of this study indicated that more caution will be needed for selecting the GCMs and using the results of the climate change analysis.

• 한국정밀공학회지
• /
• 제27권4호
• /
• pp.40-45
• /
• 2010

Evaluation of uncertainty is an ongoing process that can consume time and resources. It can also require the service of someone who is familiar with data analysis techniques. Therefore, it is important for laboratory personnel who are approaching uncertainty analysis for the first time to be aware of the resources required. International inclination of measurement filed to guarantee the traceability and confidence of measurement results discards the error concept and instead analyzes the measurement uncertainty. In this paper, we analyzed the elements of measurement uncertainty on surface roughness test which are the important things in mechanical parts test. Repeat the test by 3 men, the measurement uncertainty could be calculated.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1993년도 춘계공동학술대회 발표논문 및 초록집; 계명대학교, 대구; 30 Apr.-1 May 1993
• /
• pp.329-338
• /
• 1993

The main objective of this paper is analyzing the effects of task uncertainty on EUC characteristics and end user satisfaction. Task uncertainty were identified as an important determinant of EUC characteristics. And the moderating effect of task uncertainty on the relationship between EUC characteristics and end user satisfaction was suggested. A field study was undertaken to test the hypothesized relationships among task uncertainty, EUC characteristics, and end user satisfaction Data were collected from 138 end-user of 19 Korean business organizations. The empirical results indicated that task uncertainty was significantly related to EUC characteristics and that task uncertainty had significant effect on the relationship between EUC characteristics and end user satisfaction. Implications and future research directions are drawn for the management of EUC and for further research on EUC.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제4권1호
• /
• pp.31-36
• /
• 2015

많은 불확실성 항목들은 소프트웨어 실행 시에 존재한다. 특히, 유사한 분야에서 전반적인 시스템에 많은 영향을 주게 된다. 불확실성 요소와 같은 모호성 요소의 관리는 소프트웨어 신뢰성을 위하여 중요한 요소가 된다. 따라서 모호성을 식별하여 불확실성 요소를 제거하도록 판단기준을 프로세스화 하였다. 또한 본 연구에서는 소프트웨어 개발 시, 모호성을 제거하기 위한 불확실성 식별 기준과 불확실성 처리 프로세스 및 정량적인 평가에 대하여 제시한다.

• 품질경영학회지
• /
• 제45권3호
• /
• pp.533-547
• /
• 2017

Purpose: This study proposes the main sources of uncertainty and uncertainty analysis of a measurement system of insensitive munitions tests. Methods: We established the mathematical model for calculating measurement uncertainty of insensitive munitions tests, conducted experiments for calculating uncertainties of dynamic sensitivity and overshoot value, and estimated the distributions of uncertainty factors. Results: The measurement uncertainty calculation methods are presented, which include experimental data processing methods for calculating uncertainties of dynamic sensitivity and overshoot value. Conclusion: The measurement of explosion pressure in insensitive munitions tests is an important issue to the reporting test results and classifying reaction types. The more efforts to ensure the reliability of the insensitive munitions tests results are required.

• Journal of Information Technology Applications and Management
• /
• 제24권4호
• /
• pp.13-24
• /
• 2017

This study aims to analyze which subsidiary configuration strategy is more effective under uncertainty especially technology base multinational corporations (henceforth MNCs). In previous studies real option theory scholars argue that high breadth subsidiary configuration is most effective strategy because that provides flexibility to MNCs global network. In this study I want unveil more various types of uncertainty such as technology and learning uncertainty which are more important for technology base firm and further more examine the effect of MNCs subsidiary configuration on firm R&D performance each uncertainty case. Empirical study is performed by negative binominal model based on Japanese 108 multinational corporations. The result shows that under technology uncertainty, high breadth subsidiary configuration is better for firm R&D performance but under learning uncertainty high depth subsidiary configuration is better. Thus, the effects of MNCs subsidiary configuration on firm value can differ by types of uncertainty.

• East Asian Economic Review
• /
• 제28권1호
• /
• pp.95-140
• /
• 2024

This study empirically investigates the international transmission of China's uncertainty shocks. It estimates a time-varying parameter Bayesian global structural vector autoregressive model (TVP-BGVAR) using time series data for 33 countries to evaluate heterogeneous international linkage across countries and time. Uncertainty shocks are identified via sign restrictions. The empirical results reveal that an increase in uncertainty in China negatively affects the global economy, but those effects significantly vary over time. The effects of China's uncertainty shocks on the global economy have been significantly altered by China's WTO accession, the global financial crisis, and the recent US-China trade conflict. Furthermore, the effects of China's uncertainty shocks, typically on inflation, differ significantly across countries. Moreover, Trade openness appears crucial in explaining heterogeneous GDP responses across countries, whereas the international dimension of monetary policy appears to be important in explaining heterogeneous inflation responses across countries.