국내논문정보

이지호,Lee, Ji-Ho 대한산업보건협회 2000 산업보건 Vol.142 No.-

영 과잉 포아송 모형에 대한 베이지안 방법 연구

이지호,최태련,우윤성,Lee, Ji-Ho,Choi, Tae-Ryon,Wo, Yoon-Sung 한국통계학회 2011 응용통계연구 Vol.24 No.4

In this paper, we consider Bayesian approaches to zero inflated Poisson model, one of the popular models to analyze zero inflated count data. To generate posterior samples, we deal with a Markov Chain Monte Carlo method using a Gibbs sampler and an exact sampling method using an Inverse Bayes Formula(IBF). Posterior sampling algorithms using two methods are compared, and a convergence checking for a Gibbs sampler is discussed, in particular using posterior samples from IBF sampling. Based on these sampling methods, a real data analysis is performed for Trajan data (Marin et al., 1993) and our results are compared with existing Trajan data analysis. We also discuss model selection issues for Trajan data between the Poisson model and zero inflated Poisson model using various criteria. In addition, we complement the previous work by Rodrigues (2003) via further data analysis using a hierarchical Bayesian model. 본 논문에서는 영 과잉 계수형 자료 분석을 위한 모형중의 하나인 영 과잉 포아송 모형의 베이지안 접근 방법에 대해서 연구한다. 구체적으로는 베이지안 영 과잉 포아송 모형의 적합을 위한 사후 표본을 추출하는데 있어서, 깁스 표집기(Gibbs sampler)를 이용하는 마르코프 연쇄 몬테칼로(MCMC) 방법과 역 베이즈공식(IBF)에 의한 표본추출 방법 두 가지를 고려한다. 이러한 두 가지 사후 표본 추출방법을 비교 설명하고, IBF를 통한 사후표본을 깁스 표집기 사후표본의 수렴성 여부를 확인하는 방식에 대해서도 소개한다. 이를 바탕으로 베이지안 영 과잉 포아송 모형을 Trajan이라는 사과 품종의 발아자료(Trajan data, Marin 등, 1993)에 적용하고 모수에 대한 사후추론을 실시하고 기존의 결과와 비교한다. 또한 주어진 자료에 대하여 영 과잉 포아송 모형이 적합한지에 대한 여부를 여러 가지 모형선택 기준을 통해서 살펴보고, 아울러 기존의 자료 분석 결과 (Rodrigues, 2003)를 보완하기 위하여 계층적 베이지안 모형과 같은 대안에 대해서도 논의해본다.

모수 절약 주기적 자기회귀 모형에 관한 연구

이지호,성병찬,Lee, Jiho,Seong, Byeongchan 한국통계학회 2016 응용통계연구 Vol.29 No.1

본 논문에서는 주기적 자기회귀(periodic autoregressive) 모형에서 모수의 수를 줄이기 위한 모수 절약 주기적 자기회귀 모형을 연구하였다. 제안된 모수 절약 모형은 실증분석에서 실업률을 이용하여 기존의 계절 시계열 모형과 비교를 통하여 그 성능을 평가하였다. 모수 절약 구조를 부여하기 위하여 계절성에서 공통된 패턴을 찾아내는 방법을 사용하였으며 기존 주기적 자기회귀 모형과의 통계적 차이 유무는 LR 검정을 통해 확인하였다. 그 결과, 중장기적으로 주기적 자기회귀 모형이 기존의 계절시계열 모형보다 우수한 예측성능을 보였으며, 특히 모수 절약 주기적 자기 회귀 모형의 사용은 기존의 주기적 자기회귀 모형보다 우수한 예측성능을 나타내는 것을 확인하였다. This paper proposes a parsimonious periodic autoregressive (PAR) model. The proposed model performance is evaluated through an analysis of Korean unemployment rate series that is compared with existing models. We exploit some common features among each seasonality and confirm it by LR test for the parsimonious PAR model in order to impose a parsimonious structure on the PAR model. We observe that the PAR model tends to be superior to existing seasonal time series models in mid- and long-term forecasts. The proposed parsimonious model significantly improves forecasting performance.

다목적 최적화 기법을 이용한 수위관측망 평가

이지호,주홍준,김형수,전환돈,Lee. JiHo,Joo. HongJun,Kim. HungSoo,Jun. HwanDon 한국방재학회 2014 한국방재학회논문집 Vol.14 No.4

본 연구에서는 남강댐 유역을 대상으로 수위관측소의 수문정보량과 설치목적별 중요도를 고려하여 수위관측망을 평가하였다. 엔트로피 이론을 적용하여 산정된 수문정보량과 설치목적별 중요도를 동시에 고려하기 위해 유클리디언 거리를 이용하여 2개의 목적함수를 통합하였으며, 이를 다목적 최적화 기법을 이용하여 최적관측망을 선정하였다. 최적 수위관측망을 구축하기 위해 대표단위도를 실제 강우-유출 해석을 결과를 바탕으로 유도한 후 확률밀도함수로의 전이를 통해 엔트로피 이론에 적용하였다. 아울러 설치목적별 중요도는 MLIT(2010)에서 조사한 수위관측소별 설치목적을 참조하여 등급을 정량화 하였다. 이를 남강댐에 유역에 적용한 결과 다목적 최적화 기법을 이용하여 수위관측망을 평가한 경우가 엔트로피 이론만을 적용했을 때보다 관측소의 설치목적을 보다 잘 만족함을 확인하였다. 이는 수위관측소의 설치목적별 중요도와 자료의 특성이 동시에 반영된 결과이다. 아울러 대상유역인 남강댐 유역에서는 총 12개의 수위관측소 중 8개의 수위관측소가 필요한 것으로 판단된다. This study has evaluated the stream gauge network which is considered the total hydrological information and the importance class of stream gauge stations, and is applied to Namgang Dam basin. The total hydrological information and the importance class of stream gauge stations are two different physical parameters. We adopt the Euclidean distance method to consider them simultaneously. An equation obtained by the Euclidean distance method is used as the objective function in multiobjective optimization method to determine the optimal combination of the stream gauge stations for satisfying the both criteria: the maximum total amount of the hydrological information and the importance class of stream gauge stations. To evaluate the optimal stream gauge network, the representative unit hydrograph which is derived by real rainfall-runoff analysis is transferred into probability density function for the application of entropy theory concept. The importance class of stream gauge stations is determined by referring the result of MLIT (2010). As result of applying to Namgang Dam basin, The optimal stream gauge network obtained by the suggested method shows more satisfied the importance of stream gauge station purpose than the optimal stream gauge network obtain by the entropy theory only. This result implies that the optimal steam gauge network by the suggested method can consider both the total hydrological information and the importance class of stream gauge station. It is found that Namgang Dam basin needs at least 8 steam gauge stations among 12 steam gauge stations.

울산지역 초등학생의 알레르기비염 유병과 면역반응과의 상관성: 환자-대조군 연구

이지호,오인보,김아라,김민호,심창선,김양호,Lee, Jiho,Oh, Inbo,Kim, Ahra,Kim, Minho,Sim, Chang sun,Kim, Yangho 한국환경보건학회 2015 한국환경보건학회지 Vol.41 No.4

Objectives: This study aims to investigate the correlations between the prevalence of allergic rhinitis (AR) and cytokines among elementary school children in an industrial city, Ulsan, South Korea, and to identify major environmental risk factors associated with AR prevalence. Methods: We conducted a case-control study in June 2009 and February 2010 in order to evaluate the relationship between AR and related cytokines. Data on physician-treated prevalence over the past 12 months and potential risk factors for AR were compiled through a questionnaire from a survey of 339 schoolchildren living in different urban environments. Logistic regression analysis was carried out with propensity score matched data (n=180) to assess the influences of cytokines (IL-13, IL-33, IL-4 and IL-5) on AR prevalence and to determine which environmental factors affected AR. Results: In univariate analysis, the AR prevalence was influenced by family history of AR (mother and siblings), environmental factors (odor condition and irritated symptoms of air pollution), and indoor allergens (D. farinae and D. pteronyssinus). The t-test demonstrated that eosinophils, Immunoglobulin E (IgE), and interleukins (IL-13 and IL-5) were statistically significantly different according to treatment of allergic rhinitis over the preceeding 12 months. The results of the multiple logistic regression analysis showed that a statistically significant association between several factors (such as irritated symptoms of air pollution (OR 4.075, CI 1.735-9.568), IL-13 (OR 0.825, CI 0.734-0.928), odor condition (OR 2.409, CI 0.908-6.389), and AR history of siblings (OR 2.217, CI 0.999-4.921)) and the prevalence of AR was found after adjusting for confounders. Conclusion: These results suggest that AR prevalence is significantly associated with cytokine level, genetic background, and outdoor environmental factors. Although living in a polluted area and genetic background can contribute to an increased risk of childhood AR, cytokine level should be considered as an important factor in the treatment of AR in the last 12 months.

Propyl gallate에 의한 알레르기성 접촉 구순염

이지호,윤두희,이준영,김형옥 ( Ji Ho Lee,Doo Hee Yoon,Jun Young Lee,Hyung Ok Kim ) 대한피부과학회 1997 대한피부과학회지 Vol.35 No.2

Propyl gallste and other gallic acid esters are used as antioxidants in lipsticks, lip balms and salves, cosmetic creams and lotions, bakery products, edible fats and other pharmaceutical and industrial products. Propyl gallate is used widely but allergic contact dermatits from propyl gallate is rare. A 44-year-old female patient had pruritic multiple tiny erythematous papules and vesicles on the margin of her lip for a week. We found that the causative material of the allergic contact cheilitis was propyl gallate. We proved it with a patch test, provocation use test and quantitative and qualitative analysis of the lipstick. To our knowledge, this is the first case report of lipstick allergic contact cheilitis from propyl gallate in Korean literature. (Kor J Dermatol 1997;35(2): 374-378)

컴퓨터보조 기반 점막지지 서지컬템프레이트를 이용한 하악전치부 임플란트 식립

이지호,김성민,김명주,박정민,서미현,명훈,이종호,김명진,Lee, Jee-Ho,Kim, Soung-Min,Kim, Myung-Joo,Park, Jung-Min,Seo, Mi-Hyun,Myoung, Hoon,Lee, Jong-Ho,Kim, Myung-Jin 대한악안면성형재건외과학회 2011 Maxillofacial Plastic Reconstructive Surgery Vol.33 No.2

A 73-year-old Korean female patient with a fully edentulous mandible was planned to have five implant fixtures installed in the anterior mandible for the fixed prosthesis. After 3-dimensional (3D) computed tomographic scanning was transferred to OnDemand3D$^{(R)}$ (Cybermed Co., Seoul, Korea) software program for the virtual planning, five fixtures of MK III Groovy RP implants of Branemark System$^{(R)}$ (Nobel Biocare AB Co., Goteborg, Sweden) were installed in the anterior mandible between both mental foramens using In2Guide$^{(R)}$ (CyberMed Co., Seoul, Korea) mucosa-supported surgical template with Quick Guide Kit$^{(R)}$ (Osstem Implant Co., Seoul, Korea) systems. Fixture installations were completed successfully without any complications, such as mental nerve injury, bony bleedings, fenestrations and other unexpected events. Postoperative computed tomographic scans were aligned and fused to the planned implant, then angular and linear deviations were compared with the planned virtual implants. The mean angular deviation between the planned and actual implant axes was $3.42{\pm}1.336^{\circ}$. The mean distance between the planned and actual implant at the neck area was $0.544{\pm}0.290$ mm horizontally and $0.118{\pm}0.079$ mm vertically. The average distance between the planned and actual implant at the apex area was $1.166{\pm}0.566$ mm horizontally and $0.14{\pm}0.091$ mm vertically. These results could be considered more precise and accurate than previous reports, and even our recent results. The entire procedures of this case are reported and reviewed.

Normal Range of the Inflammation Related Laboratory Findings and Predictors of the Postoperative Infection in Spinal Posterior Fusion Surgery

이지호,이재협,Ji-Beom Kim,Hyeong-Seok Lee,Do-Yoon Lee,이동오 대한정형외과학회 2012 Clinics in Orthopedic Surgery Vol.4 No.4

Background: Inflammation related hematological parameters vary greatly depending on patients. It is not well known how much increase of which parameter warrants suspicion of postoperative infection. This study proposes to identify the normal range and the predictive factors for postoperative infection by conducting a time series analysis of the hematological parameters of patients after the spinal posterior fusion. Methods: A retrospective study was done with 608 patients who underwent spinal posterior fusion with pedicle screw fixation. Laboratory assessment including the leucocyte, neutrophil, C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR) of patients for 2 weeks after operation. The patients were divided into the one-level fusion group (group I), the two-level fusion group (group II), the three or multi-level fusion or reoperation group (group III), and the postoperative infection group (group IV). Blood was drawn before breakfast prior to the operation, and then 2-3 days, 4-7 days, 8-11 days, and 12-14 days after the operation. The leucocyte count, neutrophil count, CRP, and ESR were measured. Results: From 4-7 days after the operation, the CRP and neutrophil count of group IV were significantly higher than those of group I and II, and from 8-11 days after operation, the CRP and neutrophil counts were significantly higher than those of all groups. Twelve to fourteen days after the operation, the neutrophil count of group IV was significantly higher than that of group I and II, while the neutrophil count of group III was also higher than that of group I. The lower limit of the 95% confidence interval (CI) of the CRP and neutrophil count group IV was greater than the upper limit of the 95% CI of group I and II. The ESR of group IV was significantly higher than that of group I and III. Conclusions: If the postoperative CRP and neutrophil counts are high, or if the CRP begins to rise again 8 days after the operation, the likelihood of infection increases, but caution must be exercised in interpreting the results. If the hematological parameters are higher than the lower limit of the 95% CI of the postoperative infection group, infection must be strongly suspected.

엔트로피 방법론을 통한 관측망 평가의 한계점 논의

이지호,김수전,전환돈,Lee, Ji-Ho,Kim, Su-Jeon,Jeon, Hwan-Don 한국수자원학회 2019 물과 미래(한국수자원학회지) Vol.52 No.9

DCC 모형에서 동태적 상관계수 추정법의 효율성 비교

이지호,성병찬,Lee, Jiho,Seong, Byeongchan 한국통계학회 2015 응용통계연구 Vol.28 No.5

본 논문에서는 다변량 DCC(dynamic conditional correlation) GARCH 모형에서 동태적 상관계수를 추정하기 위한 대표적 방법인 쌍별 추정법과 다차원 추정법의 효율성을 비교한다. 이를 위하여 금융 시장의 변동성을 반영하는 다변량 시계열을 생성하고 이에 대한 DCC GARCH 모형을 수립 및 추정하는 시뮬레이션을 실시하였다. 또한 KOSPI 200 섹터지수를 이용하여 포트폴리오를 구성하고 이의 변동성 추정 및 VaR 계산을 통하여 동태적 상관계수 추정에 대한 정확성을 평가하였다. 그 결과로서, 전반적으로 다차원 추정법이 쌍별 추정법보다 우수함을 발견하였다. 특히, 다차원 추정법에서 상대적으로 상관관계가 낮은 시계열을 추가할수록 쌍별 시계열에 대한 동태적 상관계수 추정의 정확성을 높여줌을 발견하였다. We compare the performance of two representative estimation methods for the dynamic conditional correlation (DCC) GARCH model. The first method is the pairwise estimation which exploits partial information from the paired series, irrespective to the time series dimension. The second is the multi-dimensional estimation that uses full information of the time series. As a simulation for the comparison, we generate a multivariate time series similar to those observed in real markets and construct a DCC GARCH model. As an empirical example, we constitute various portfolios using real KOSPI 200 sector indices and estimate volatility and VaR of the portfolios. Through the estimated dynamic correlations from the simulation and the estimated volatility and value at risk (VaR) of the portfolios, we evaluate the performance of the estimations. We observe that the multi-dimensional estimation tends to be superior to pairwise estimation; in addition, relatively-uncorrelated series can improve the performance of the multi-dimensional estimation.