• Wind and Structures
• /
• v.30 no.2
• /
• pp.141-153
• /
• 2020

Probabilistic information regarding directional extreme wind speeds is important for the precise estimation of the design wind loads on structures. A joint probability distribution model of directional extreme typhoon wind speeds is established using Monte Carlo simulation and empirical copula function to fully consider the correlations of extreme typhoon wind speeds among the different directions. With this model, a procedure for estimating directional extreme wind speeds for given return periods, which ensures that the overall risk is distributed uniformly by direction, is established. Taking 5 typhoon-prone cities in China as examples, the directional extreme typhoon wind speeds for given return periods estimated by the present method are compared with those estimated by the method proposed by Cook and Miller (1999). Two types of directional factors are obtained based on Cook and Miller (1999) and the UK standard's drafting committee (Standard B, 1997), and the directional risks for the given overall risks are discussed. The influences of the extreme wind speed correlations in the different directions and the simulated typhoon wind speed sample sizes on the estimated extreme wind speeds for a given return period are also discussed.

• Communications for Statistical Applications and Methods
• /
• v.31 no.1
• /
• pp.65-85
• /
• 2024

The tail probability of a function of a multivariate random variable is not easy to estimate by the crude Monte Carlo simulation. When the occurrence of the function value over a threshold is rare, the accurate estimation of the corresponding probability requires a huge number of samples. When the explicit form of the cumulative distribution function of each component of the variable is known, the inverse transform likelihood ratio method is directly applicable scheme to estimate the tail probability efficiently. The method is a type of the importance sampling and its efficiency depends on the selection of the importance sampling distribution. When the cumulative distribution of the multivariate random variable is represented by a copula and its marginal distributions, we develop an iterative algorithm to find the optimal importance sampling distribution, and show the convergence of the algorithm. The performance of the proposed scheme is compared with the crude Monte Carlo simulation numerically.

• Communications for Statistical Applications and Methods
• /
• v.21 no.1
• /
• pp.23-43
• /
• 2014

We study the dependence between the insureds in multiple-life insurance contracts. With the future lifetimes of the insureds modeled as correlated random variables, both premium and reserve are different from those under independence. In this paper, Gaussian copula is used to impose the dependence between the insureds with Gompertz marginals. We analyze the change of the reserves of standard multiple-life insurance contracts at various dependence levels. We find that the reserves based on the assumption of dependent lifetimes are quite different for some contracts from those under independence as its correlation increase, which elucidate the importance of the dependence model in multiple-life contingencies in both theory and practice.

• Proceedings of the KIEE Conference
• /
• 2015.07a
• /
• pp.499-500
• /
• 2015

최근, 전력소비자 측면에서의 에너지 관리 역할이 중요해짐에 따라 다양한 에너지 관리시스템들이 개발되고 있다. 이 중, 가정용 에너지 관리시스템(Home Energy Management System, HEMS)은 전기가격 및 (신뢰도)수요반응 요청 등에 따라 가전기기의 사용을 조절함으로써, 전력사용 최적화(전력사용 비용 최소화)를 수행하게 된다. 이러한 최적화 문제 내에서 각 가전기기에 대한 사용패턴(편의성)과 관련된 제약조건이 고려되어야 한다. 본 논문에서는, 다변량 간 상관관계(의존성)을 추정하고, 이를 근거로 데이터를 샘플링 하는 데에 유용한 Copula 함수를 이용하여, 각 가전기기의 사용패턴을 모델링하고, 이를 최적화 문제 내 제약조건으로 고려할 수 있는 방법에 대한 연구를 수행하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.351-351
• /
• 2017

2014-2015년 우리나라 강수량이 평년에 비해 절반수준에 미치지 못해 극심한 가뭄을 일으켰으며, 이는 댐 용량 부족, 지하수 고갈 등 다양한 피해를 발생시켰다. 특히 소양강댐의 경우 1978년 이루 두 번째로 낮은 수위를 기록한바 있다. 우리나라의 경우 가뭄은 약 2-3냔 주기로 발생하고 있으며, 특히 2015년에 겪었던 가뭄은 물 용수공금 측면에서 막대한 영향을 미친 것으로 평가되어 신뢰성 있는 가뭄 분석이 중요한 요소로 대두되고 있다. 또한 지구온난화로 인해 기후변화의 영향으로 강수량의 증가가 일반적으로 전망되지만, 상대적으로 증가된 강우변동성으로 인해 가뭄 발생 빈도 및 강도도 동시에 증가할 것으로 전망되고 있다. 이러한 이유로 본 연구에서는 현재 가뭄을 신뢰성있게 평가하기 위해 Trivariate Copula 함수를 활용하여 가뭄분석을 수행하였다. 기존연구에서는 가뭄 지속시간(drought duration), 가뭄 심도(drought severity)를 활용한 이변량 가뭄 빈도 해석을 수행하였지만, 이는 다소 과소 추정 될 개연성이 있다. 이러한 이유로 본 연구에서는 가뭄강도(drought intensity) 변량을 추가로 분석하여 Trivariate frequency analysis 기법을 개발하였으며, 서울 관측소를 대상으로 분석하였다. 분석 결과 현재 가뭄은 역대 발생했던 가뭄 중 가장 큰 빈도를 기록하여 이에 대한 효과적인 가뭄 관리체계를 마련하기 위한 기초자료를 제공할 수 있을 것으로 기대된다. 또한 기존 Bivariate 빈도해석의 경우 Trivariate 빈도해석 보다 가뭄위험도를 다소 과소추정하는 것으로 나타나 Trivariate 해석이 다소 현실적인 접근 방법이라 사료된다.

• Journal of Korea Water Resources Association
• /
• v.52 no.12
• /
• pp.985-999
• /
• 2019

Over the last decade, droughts have become more severe and frequent in many regions, and several studies have been conducted to explore the recent drought. Copula-based bivariate drought frequency analysis has been widely used to evaluate drought risk in the context of point frequency analysis. However, the relatively significant uncertainties in the parameters are problematic when available data are limited. For this reason, the primary purpose of this study is to develop a regional drought frequency model based on the Copula function. All parameters, including marginal and copula functions in the regional frequency model, were estimated simultaneously. Here, we present a case study of recent drought 2013-2015 over the Han-River watershed where severe drought risk is consistently found to increase. The proposed model provided a reliable way to significantly reduce the uncertainty of parameters with a Bayesian modeling framework. The uncertainty of the joint return period in the regional frequency analysis is nearly three times lower than that of the point frequency analysis. Accordingly, DIC values in the regional frequency analysis model are significantly decreased by 15. The results confirm that the proposed model is not only reliably representing characteristics of historical droughts and dependencies between drought variables, but also providing the efficacy of understanding regional drought characteristics.

• Journal of Digital Convergence
• /
• v.11 no.9
• /
• pp.103-111
• /
• 2013

As the demand-oriented management has been getting important in Supply Chain Management (SCM), various forecasting methods have been suggested including regression analyses. However, dependency structures among variables have been captured by a correlation coefficient, only. It results in inaccurate demand predictions. This paper suggests a new and effective forecasting modeling framework using student's t-copula function. In order to show overall modeling procedures framework, heavy tail typed numerical data and its copula estimations are provided. The suggested methodology can contribute to decrease the bullwhip effect and to stabilize volatile environment in a supply chain network.

• Journal of Korea Water Resources Association
• /
• v.47 no.2
• /
• pp.171-182
• /
• 2014

The objective of this study is to evaluate the utilization of bivariate joint drought index in South Korea. In order to develop the bivariate joint drought index, in this study, Clayton copula was used to estimate the joint distribution function and the calibration method was employed for parameter estimation. Precipitation and soil moisture data were selected as input data of bivariate joint drought index for period of 1977~2012. The time series analysis, ROC (Receiver Operating Characteristic) analysis, spatial analysis were used to evaluate the bivariate joint drought index with SPI (Standardized Precipitation Index) and SSI (Standardized Soil moisture Index). As a result, SPI performed better for drought onset and SSI for drought demise. On the other hand the bivariate joint drought index captured both drought onset and demise very well. The ROC score of bivariate joint drought index was higher than that of SPI and SSI, and it also reflected the local drought situations. The bivariate joint drought index overcomes the limitations of existing drought indices and is useful for drought analysis.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2016.05a
• /
• pp.41-41
• /
• 2016

최근 다변량 확률모형을 이용한 빈도해석이 수문자료 등에 적용되면서 다양하게 연구되고 있으며 다변량 확률모형 중 copula 모형은 주변분포형에 대한 제약이 없어 여러 분야에 걸쳐 활발히 연구되고 있다. 강우자료는 기존 일변량 빈도해석을 수행하기 위하여 사용하던 block maxima 방법 대신 최소무강우시간(inter event time)을 통하여 강우사상을 추출하여 표본으로 사용한다. 또한 기후변화로 인한 강우량의 변화등에 대응하기 위하여 비정상성 Generalized Extreme Value(GEV)와 Gumbel 등의 확률분포형에 대한 연구도 많은 부분 이루어져 있다. 본 연구에서는, Archimedean copula 모형을 이용하여 이변량 확률모형을 구축하면서 여기에 사용되는 주변분포형에 정상성/비정상성 분포형을 적용하였다. 모형의 매개변수는 inference function for margin 방법을 이용하였으며 주변분포형으로는 정상성/비정상성 GEV, Gumbel 모형을 적용하였다. 결과로 정상성/비정상성 경향을 나타내는 지점을 구분하고 각 지점에 대한 정상성/비정상성 주변분포형을 적용한 이변량 확률분포형을 구하였다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.305-305
• /
• 2022

A better approach for assessing meteorological drought occurrences is increasingly important in mitigating and adapting to the impacts of climate change, as well as strategies for developing early warning systems. The present study defines meteorological droughts as a period with an abnormal precipitation deficit based on monthly precipitation data of 18 gauging stations for the Han River watershed in the past (1974-2015). This study utilizes a Bayesian parameter estimation approach to analyze the effects of climate change on future drought (2025-2065) in the Han River Basin using the Coupled Model Intercomparison Project Phase 6 (CMIP6) with four bias-corrected general circulation models (GCMs) under the Shared Socioeconomic Pathway (SSP)2-4.5 scenario. Given that drought is defined by several dependent variables, the evaluation of this phenomenon should be based on multivariate analysis. Two main characteristics of drought (severity and duration) were extracted from precipitation anomalies in the past and near-future periods using the copula function. Three parameters of the Archimedean family copulas, Frank, Clayton, and Gumbel copula, were selected to fit with drought severity and duration. The results reveal that the lower parts and middle of the Han River basin have faced severe drought conditions in the near future. Also, the bivariate analysis using copula showed that, according to both indicators, the study area would experience droughts with greater severity and duration in the future as compared with the historical period.