• 한국안전학회지
• /
• 제26권5호
• /
• pp.130-135
• /
• 2011

This study deals with the accident models of arterial road sections. The objectives is to develop the models by number of lane. In pursuing the above, this study gives particular emphasis to dividing the 474 small link sections, collecting the accident data of 2007, and applying the statistical programs of SPSS17.0 and NLOGIT4.0. The main results are as follows. First, the number of accidents of two-lane roads were analyzed to be 59.9% of totals and to be the most of all. Second, one Poisson and two negative binomial regression models which were all statistically significant were developed. Finally, the common variables of all models were evaluated to be ADT and number of exit/entry which were all positive to the accidents.

• 응용통계연구
• /
• 제31권6호
• /
• pp.733-749
• /
• 2018

어떠한 연구에서 관심의 대상이 되는 관찰치가 부분적으로 관측 가능할 때 표본선택의 문제가 일어난다. 이러한 자료를 분석하기 위해 헤크만은 표본선택 모형을 개발하였고 이변량 정규분표의 가정 하에 최대우도방법을 사용하여 모수를 추정하였다. 최근 이항자료와 포아송 자료에 대한 표본선택모형이 제안되었다. 이를 분포조정에 기초하여 과대산포 자료에 대한 모형으로 확장하고자 한다. 표본선택이 없는 과대산포 자료는 흔히 음이항 분포로 분석되어진다. 따라서 음이항 분포를 이용하고 분포조정을 도입한 과대산포 자료에 대한 새로운 모형을 제시하고자 한다. 실제 자료를 이용하여 분석을 하였다. 모의실험 결과 프로파일 우도함수를 이용하여 모수에 대해 추정한 결과는 안정적이다.

• 한국ITS학회:학술대회논문집
• /
• 한국ITS학회 2007년도 제6회 추계학술대회 및 정기총회
• /
• pp.92-99
• /
• 2007

• 한국도로학회논문집
• /
• 제11권2호
• /
• pp.151-158
• /
• 2009

본 연구는 지방부에서의 후미추돌사고를 다루고 있다. 이 연구의 목적은 신호교차로의 후미주돌 사고특성을 분석하고 청주 청원에 대한사고모형을 개발하는 것이다. 이를 위해, 이 연구에서는 도시부와 지방부의 특성을 비교하는데 중점을 두고 있다. 이 연구에서 사용된 종속변수는 사고건수와 EFDO(equivalent property damage only)이며, 독립변수는 교통량과 기하구조 요소들로 이루어졌다. 주요 연구결과는 다음과 같다. 첫째, 사고건수를 종속변수로 이용한 포아송 회귀모형과 EFDO를 종속변수로 이용한 음이항 회귀모형이 통계적으로 적합한 것으로 분석된다. 둘째, 포아송 회귀분석 결과 나타난 독립변수들은 중차량비, 교통량 합계 그리고 차량 유출입구 합계이며 음이항 회귀분석으로 나타난 요인은 주도로 폭, 교통량 합계 그리고 중차량비로 분석된다. 마지막으로, 지방부에서의 특정 독립변수는 주도로 폭과 중차량비 그리고 차량 유출입구 합계이다.

• 보건행정학회지
• /
• 제25권4호
• /
• pp.285-294
• /
• 2015

Background: Suicide is one of important health problems in Korea. Previous studies showed factors associated with suicide in individual levels. However, suicide was influenced by society that individuals belong to, so it was required to analyze suicide in local levels. The purpose of this study was to analyze the regional disparities of suicide mortality by gender and the association between local characteristics and suicide mortality. Methods: This study included 229 city county district administrative districts in Korea. Age- and sex-standardized suicide mortality and age-standardized suicide mortality (male/female) were used as dependent variables. City county district types, socio-demographics (number of divorces per 1,000 population, number of marriages per 1,000 population, and single households), financial variable (financial independence), welfare variable (welfare budget), and health behavior/status (perceived health status scores and EuroQol-5 dimension [EQ-5D]) were used to represent the local characteristics. We used hot-spot analysis to identify the spatial patterns of suicide mortality and negative binomial regression analysis to examine factors affecting suicide mortality. Results: There were differences in distribution of suicide mortality and hot-spot regions of suicide mortality by gender. Negative binomial regression analysis provided that city county district types (city), number of divorces per 1,000 population, financial independence, and EQ-5D had significant influences on the age- and sex-standardized suicide mortality per 100,000. Factor influencing suicide mortality was the number of divorces per 1,000 population in both male and female. Conclusion: Study results provided evidences that suicide mortality among regions was differed by gender. Health policy makers will need to consider gender and local characteristics when making policies for suicides.

• 한국ITS학회 논문지
• /
• 제17권4호
• /
• pp.14-25
• /
• 2018

최근 빅데이터 시대의 도래와 함께 교통사고와 관련된 요인을 설명하기 용이해졌다. 이에따라 최신 분석 기법을 적용하여 교통사고 자료를 분석하고 시사점을 도출할 필요가 있다. 본 연구의 목적은 고속도로 교통사고 자료를 이용하여 고속도로의 주요 분석 단위인 콘존의 교통사고 건수를 예측하기 위하여 음이항 회귀모형과 딥 러닝을 이용한 기법을 적용하고 예측 성능을 비교하였다. 예측 성능 비교 결과, 딥 러닝 모형의 MOE들이 음이항 회귀모형에 비해 다소 우수한 것으로 나타났으나, MAD 기준으로 차이는 미미한 것으로 나타났다. 하지만 딥 러닝을 이용할 경우 다른 독립변수들을 추가하는 것이 용이하고, 모형의 구조 등을 변경할 경우 예측 신뢰도를 더욱 증가시킬 수 있을 것으로 판단된다.

• 보건행정학회지
• /
• 제26권4호
• /
• pp.343-351
• /
• 2016

Background: This study aims to analyze the effects of insurance types on the medical service uses for heart failure inpatients using propensity score matching (PSM). Methods: 2014 National inpatient sample based on health insurance claims data was used in the analysis. PSM was applied to control factors influencing the service uses except insurance types. Negative binomial regression was used after PSM to analyze factors that had influences on the service uses among inpatients. Subjects were divided by health insurance type, national health insurance (NHI) and medical aid (MA). Total charges and length of stay were used to represent the medical service uses. Covariance variables in PSM consist of sociodemographic characteristics (gender, age, Elixhauser comorbidity index) and hospital characteristics (hospital types, number of beds, location, number of doctors per 50 beds). These variables were also used as independent variables in negative binomial regression. Results: After the PSM, length of stay showed statistically significant difference on medical uses between insurance types. Negative binomial regression provided that insurance types, Elixhauser comorbidity index, and number of doctors per 50 beds were significant on the length of stay. Conclusion: This study provided that the service uses, especially length of stay, were differed by insurance types. Health policy makers will be required to prepare interventions to narrow the gap of the service uses between NHI and MA.

• 한국안전학회지
• /
• 제29권5호
• /
• pp.154-159
• /
• 2014

This study deals with the pedestrian accidents of intersections in case of Cheongju. The objective is to develop the pedestrian accident models using Tobit regression model. In pursuing the above, the pedestrian accident data from 2007 to 2011 were collected from TAAS data set of Road Traffic Authority. To analyze the accident, Poisson, negative binomial and Tobit regression models were utilized in this study. The dependent variable were the number of accident by intersection. Independent variables are traffic volume, intersection geometric structure and the transportation facility. The main results were as follows. First, Tobit model was judged to be more appropriate model than other models. Also, these models were analyzed to be statistically significant. Second, such the main variables related to accidents as traffic volume, pedestrian volume, number of traffic island, crossing length and the pedestrian countdown signal systems were adopted in the above model.

• 한국도로학회논문집
• /
• 제20권2호
• /
• pp.67-74
• /
• 2018

PURPOSES : The goal of this study is the development of roundabout accident models for urban and non-urban areas. METHODS : This study performed a comparative analysis of the regional factors affecting accidents. Traffic accident data were collected for the period 2010~2014 from the TAAS data set of the Road Traffic Authority. To develop the roundabout accident models, the Poisson and negative binomial regression models were used. A total of 25 explanatory variables such as geometry, and traffic volume were used. RESULTS : The key findings are as follows: First, it was found that the null hypotheses that the number of accidents is the same should be rejected. Second, three Poisson regression accident models, which are statistically significant (${\rho}^2$ of 0.154 and 0.385) were developed. Third, it was noted that although the common variable of the three models (models I~III) is the number of entry lanes, the specific variables are entry lane width, roundabout sign, number of circulatory roadways, splitter island, number of exit lanes, exit lane width, number of approach roads, and truck apron. CONCLUSIONS : The results of this study can provide suggestive countermeasures for decreasing the number of roundabout accidents.

• 응용통계연구
• /
• 제21권4호
• /
• pp.603-613
• /
• 2008

셀 수 있는 이산 자료 중에서 일반적인 모형에 비하여 영의 빈도가 과도하게 많이 관측되는 자료가 있다. 이러한 경우에 포아송 또는 음이항회귀모형과 같은 일반적인 회귀모형에 의한 분석은 적절하지 못하다. 본 논문에서는 영과잉 포아송회귀모형과 영과잉 음이항회귀모형에 대하여 베이지안 분석을 하였다. 또한, 마코브 연쇄 몬테카롤로 방법으로 계산한 베이즈 요인을 이용하여 모형선택을 하였다. 실제 교통사고 자료를 분석하여 이론적인 결과들을 뒷받침하였다.