• 한국도로학회논문집
• /
• 제14권5호
• /
• pp.133-143
• /
• 2012

PURPOSES: Using the collected data for crash, traffic volume, and design elements on ramps between 2007 and 2009, this research effort was initiated to develop traffic crash prediction models for expressway ramps. METHODS: Three negative binomial regression models and three zero-inflated negative binomial regression models were developed for individual ramp types, including direct, semi-direct and loop, respectively. For validating the developed models, authors compared the estimated crash frequencies with actual crash frequencies of twelve randomly selected interchanges, the ramps of which have not been used for model developing. RESULTS: The results show that the negative binomial regression models for direct, semi-direct and loop ramps showed 60.3%, 63.8% and 48.7% error rates on average whereas the zero-inflated negative binomial regression models showed 82.1%, 120.4% and 57.3%, respectively. CONCLUSIONS: Conclusively, the negative binomial regression models worked better in traffic crash prediction than the zero-inflated negative binomial regression models for estimating the frequency of traffic accidents on expressway ramps.

• 한국도로학회논문집
• /
• 제18권4호
• /
• pp.83-92
• /
• 2016

PURPOSES : The purpose of this study was to develop safety performance functions (SPFs) that use zero-inflated negative binomial regression models for urban intersections in central business districts (CBDs), and to compare the statistical significance of developed models against that of regular negative binomial regression models. METHODS : To develop and analyze the SPFs of intersections in CBDs, data acquisition was conducted for dependent and independent variables in areas of study. We analyzed the SPFs using zero-inflated negative binomial regression model as well as regular negative binomial regression model. We then compared the results by analyzing the statistical significance of the models. RESULTS : SPFs were estimated for all accidents and injury accidents at intersections in CBDs in terms of variables such as AADT, Number of Lanes at Major Roads, Median Barriers, Right Turn with an Exclusive Turn Lane, Turning Guideline, and Front Signal. We also estimated the log-likelihood at convergence and the likelihood ratio of SPFs for comparing the zero-inflated model with the regular model. In he SPFs, estimated log-likelihood at convergence and the likelihood ratio of the zero-inflated model were at -836.736, 0.193 and -836.415, 0.195. Also estimated the log-likelihood at convergence and likelihood ratio of the regular model were at -843.547, 0.187 and -842.631, 0.189, respectively. These figures demonstrate that zero-inflated negative binomial regression models can better explain traffic accidents at intersections in CBDs. CONCLUSIONS : SPFs that use a zero-inflated negative binomial regression model demonstrate better statistical significance compared with those that use a regular negative binomial regression model.

• 지역연구
• /
• 제31권3호
• /
• pp.39-54
• /
• 2015

본 연구는 서울시에서 추진한 청계천 복원사업에 대한 경제적 가치를 평가하기 위해 심층출구면접조사 방식으로 수집된 자료를 바탕으로 여행비용법(Travel Cost Method, TCM)을 적용하였다. 가산자료의 특성을 감안하여 분석모형은 포아송모형(Poisson Model, PM), 음이항모형(Negative Binomial, NB), 절단된 포아송모형(Zero-truncated Poisson, ZTP), 그리고 절단된 음이항모형(Zero-truncated Negative Binomial, ZTNB)을 사용하였다. 분석결과 추정계수들은 통계적으로 유의하게 나타났고 일반적인 소비자경제이론에 부합하는 결과가 도출되었다. 조사된 자료에서 과산포현상(Over-dispersion)이 발견되었으며 모형적합도검정을 통해서 절단된 음이항모형(Zero-truncated Negative Binomial, ZTNB)이 청계천 방문객의 수요를 추정하는 데 최적모형으로 선정되었다. 생태하천복원사업인 청계천복원사업의 경제적 가치를 추정하기 위해 방문객의 연평균 방문횟수와 최적모형에서 추정된 계수를 통해서 분석한 결과 청계천의 경제적 가치는 2013년 기준으로 연간 약 1,902 원으로 추정되었다.

• 자원ㆍ환경경제연구
• /
• 제23권2호
• /
• pp.331-347
• /
• 2014

본 연구는 서해에 위치한 태안군을 방문하는 유어객의 방문 편익을 분석하기 위하여 개별여행비용법과 가산자료모형을 이용하여 수요모형을 추정하고 경제적 가치를 측정하였다. 여행객방문속성을 감안하여 분석모형으로 포아송모형(Poisson Model, PM), 음이항모형(Negative Binomial, NB), 절단된 포아송모형(Truncated Poisson Model, TPM), 절단된 음이항모형(Truncated Negative Binomial, TNB)을 분석에 이용하였다. 분석결과 추정계수들은 통계적으로 유의하게 나타났으며, 경제이론에도 부합되는 결과를 도출하였다. 과산포(overdispersion)현상은 발견되지 않았으며, 모형적합도검정을 통하여 절단된 포아송모형이 적정모형으로 선정되었다. 또한, 서해 태안군 해역내 유어활동의 경제적가치를 추정하기 위하여 유어객의 연평균 출조횟수와 최적모형으로 선정된 절단된포아송모형(TPM)으로 분석한 결과 1인 1회 출조당 경제적 가치는 254,453원이며, 1인당 연간 총 경제적 가치는 1,536,896원으로 도출되었다.

• Communications for Statistical Applications and Methods
• /
• 제19권6호
• /
• pp.809-817
• /
• 2012

We propose a semiparametric inference for a generalized varying coefficient partially linear model(VCPLM) for negative binomial data. The VCPLM is useful to model real data in that varying coefficients are a special type of interaction between explanatory variables and partially linear models fit both parametric and nonparametric terms. The negative binomial distribution often arise in modelling count data which usually are overdispersed. The varying coefficient function estimators and regression parameters in generalized VCPLM are obtained by formulating a penalized likelihood through smoothing splines for negative binomial data when the shape parameter is known. The performance of the proposed method is then evaluated by simulations.

• Journal of the Korean Data and Information Science Society
• /
• 제24권4호
• /
• pp.895-900
• /
• 2013

In this research, we propose a simple bivariate zero inflated negative binomial regression model with different dispersion for bivariate count data with excess zeros. An application to the demand for health services shows that the proposed model is better than existing models in terms of log-likelihood and AIC.

• Journal of the Korean Data and Information Science Society
• /
• 제17권4호
• /
• pp.1405-1412
• /
• 2006

We shall derive the UMVUE of the tail probability in Poisson, Binomial, and negative Binomial distributions, and compare means squared errors of the UMVUE and the MLE of the tail probability in each case.

• 수산경영론집
• /
• 제38권2호
• /
• pp.79-101
• /
• 2007

The purpose of this study is to estimate the economic value of the Songieong Beach in Off-season, using a Individual Travel Cost Model(ITCM). Songieong Beach is located in Busan but far away from city. These days, however, the increased rate of traffic inflow to the Songieong beach and the five-day working week are reflected in the trend analysis. Moreover, people have changed psychological value. For that reason, visitors are on the increase on the beach in off-season. The ITCM is applied to estimate non-market value or environmental Good like a Contingent Valuation Method and Hedonic Price Model etc. The ITCM was derived from the Count Data Model(i.e. Poisson and Negative Binomial model). So this paper compares Poisson and negative binomial count data models to measure the tourism demands. The data for the study were collected from the Songjeong Beach on visitors over the a week from November 1 through November 23, 2006. Interviewers were instructed to interview only individuals. So the sample was taken in 113. A dependent variable that is defined on the non-negative integers and subject to sampling truncation is the result of a truncated count data process. This paper analyzes the effects of determinants on visitors' demand for exhibition using a class of maximum-likelihood regression estimators for count data from truncated samples, The count data and truncated models are used primarily to explain non-negative integer and truncation properties of tourist trips as suggested by the economic valuation literature. The results suggest that the truncated negative binomial model is improved overdispersion problem and more preferred than the other models in the study. This paper is not the same as the others. One thing is that Estimating Value of the Beach in off-season. The other thing is this study emphasizes in particular 'travel cost' that is not only monetary cost but also including opportunity cost of 'travel time'. According to the truncated negative binomial model, estimates the Consumer Surplus(CS) values per trip of about 199,754 Korean won and the total economic value was estimated to be 1,288,680 Korean won.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제11권2호
• /
• pp.15-29
• /
• 2007

In this paper, we explore some interesting models of the quasi-negative-binomial distribution based on difference differential equations applicable to theory of microorganisms and the situations like that. Some characterizations based on conditional distributions and damage process have been obtained. Further, the distribution of number of accidents as the quasi-negative-binomial distribution in the light of Irwin's theory of ";proneness-liability"; model has been derived. Finally, the proposed model (QNBD) has been applied to study the Shunting accidents, home injuries, and strikes in industries.

• International journal of advanced smart convergence
• /
• 제10권4호
• /
• pp.215-224
• /
• 2021

The purpose of this study is to estimate the factors that affect college students' drinking needs and spending. An analysis model to estimate the determinants affecting drinking needs was applied with a truncated Poisson model and a truncated negative binomial model. Tests to select more appropriate models of the two types were made using the comparison of log-likelihood function and the over-dispersion test. The analysis result was interpreted by applying the truncated negative binomial model as the truncated Poisson model showed over-dispersion. We also applied the Tobit model to analyze the determinantsthat affect college students' expenditure on drinking. According to the analysis, gender, grade, allowance and parental occupation were the factors influencing statistics, and gender, type of household income, and student religion were the factors influencing expenditure.