• Proceedings of the Korea Water Resources Association Conference
• /
• 1992.07a
• /
• pp.435-442
• /
• 1992

• Proceedings of the Korea Water Resources Association Conference
• /
• 1981.08a
• /
• pp.37-41
• /
• 1981

• The Korean Journal of Applied Statistics
• /
• v.25 no.1
• /
• pp.67-79
• /
• 2012

The car insurance market in Korea has already entered (or is in the process of entry) a mature market that is characterized by increased competition by market participants. Participants are expected to compete more intensively in order to survive. Together with a slowdown in market growth the goal of non-life insurers' marketing strategies is to enhance existing customer loyalty because it is easier to raise their loyalty via customer satisfaction than to attract new customers in a stagnant market. In this article, we investigate what factors affect customer loyalty, and suggest some specific ways to establish and implement marketing strategies. We use a generalized additive partial linear model in order to find some significant factors.

• The Korean Journal of Applied Statistics
• /
• v.30 no.5
• /
• pp.775-791
• /
• 2017

We frequently encounter binary data in real life. Logistic, Probit, Cauchit, Complementary log-log models are often used for binary data analysis. In order to analyze binary data, Liu (2004) proposed a Robit model, in which the inverse of cdf of the Student's t distribution is used as a link function. Kim et al. (2008) also proposed a generalized t-link model to make the binary regression model more flexible. The more flexible skewed distributions allow more flexible link functions in generalized linear models. In the sense, we propose a binary data regression model using skewed generalized t distributions introduced in Theodossiou (1998). We implement R code of the proposed models using the glm function included in R base and R sgt package. We also analyze Pima Indian data using the proposed model in R.

• The Korean Journal of Applied Statistics
• /
• v.17 no.2
• /
• pp.185-196
• /
• 2004

It is an important research area in Integrated Pest Management System to estimate the pest density within plants, because the artificial controls such as spraying pesticides or biological enemies depend on the information of pest density. This paper studies the population density distribution of two-spotted spider mite in glasshouse roses. As the data were collected repeatedly on the same subject, Subject-Specific and Population Averaged approaches are used and compared.

• The Korean Journal of Applied Statistics
• /
• v.21 no.4
• /
• pp.665-674
• /
• 2008

Most of existing studies about the small area estimation deal with the estimation of parameters based on cross-sectional data. However, since many official statistics are repeatedly collected at a regular interval of time, for instance, monthly, quarterly, or yearly, we need an alternative model which can handle characteristics of these kinds of data. In this paper, we investigate the generalized estimating equation which can model time-dependency among response variables and is useful to analyze repeated measurement or longitudinal data. We compare with the generalized linear model and the generalized estimating equation through the estimation of unemployment rates of 25 areas in Gyeongsangnam-do and Ulsan. The data consist of the status of employment and some covariates from January to December 2005.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.1
• /
• pp.101-109
• /
• 2015

SMRs (standardized mortality rates) for major diseases, accidents, cancer are considered in small areas of administrative units such as Eup/Myeon/Dong from years 2005 to 2008. Due to small sample issue in small areas, the precision of directly estimated crude SMR for each area can be low. In this study, we consider the HGLM (hierarchical generalized linear model) with MRF (Markov random field) to account for the spatial correlations among the small areas. The effects of covariates for cause of mortality by Dongs in Seoul and disease maps based on the estimated SMR are presented. The results suggest how we analyze and interpret the difference in mortalities by small areas such as Dongs by revealing the spatial patterns.

• Korean Journal of Agricultural Science
• /
• v.7 no.2
• /
• pp.131-144
• /
• 1980

The Linear Regression Model to extend the monthly runoff data in the short-recorded river was proposed by the author in 1979. Here in this study generalization precedure is made to apply that model to any given river basin and to any given station. Lengthier monthly runoff data generated by this generalized model would be useful for water resources assessment and waterworks planning. The results are as follows. 1. This Linear Regression Model which is a transformed water-balance equation attempts to represent the physical properties of the parameters and the time and space varient system in catchment response lumpedly, qualitatively and deductively through the regression coefficients as component grey box, whereas deterministic model deals the foregoings distributedly, quantitatively and inductively through all the integrated processes in the catchment response. This Linear Regression Model would be termed "Statistically deterministic model". 2. Linear regression equations are obtained at four hydrostation in Geum-river basin. Significance test of equations is carried out according to the statistical criterion and shows "Highly" It is recognized th at the regression coefficients of each parameter vary regularly with catchment area increase. Those are: The larger the catchment area, the bigger the loss of precipitation due to interception and detention storage in crease. The larger the catchment area, the bigger the release of baseflow due to catchment slope decrease and storage capacity increase. The larger the catchment area, the bigger the loss of evapotranspiration due to more naked coverage and soil properties. These facts coincide well with hydrological commonsenses. 3. Generalized diagram of regression coefficients is made to follow those commonsenses. By this diagram, Linear Regression Model would be set up for a given river basin and for a given station (Fig.10).

• The Korean Journal of Applied Statistics
• /
• v.4 no.2
• /
• pp.117-127
• /
• 1991

Coefficient of determination $R^2$ is most frequently used descriptive measure in practical use of linear regression analysis. But there have been controversies on defining this measure in the cases of linear regression without the intercept, weighted linear regression and robust linear regression. Several authors such as Kvalseth(1985) and Willet and Singer(1988) proposed many variations of $R^2$ to meet the situations. However, theire measures are not satisfactory due to the lack of a universal principle. In this study, we propose a unfied approach to defining the coefficient of determination $R^2$ using the concept of likelihood distance. This new measure is in good accordance with typical $R^2$ in linear regression and, moreover, can be applied to nonlinear regression models and generalized linear models such as logit and log-linear models.

• Proceedings of the KOR-KST Conference
• /
• 2003.02a
• /
• pp.75-93
• /
• 2003

Hub-and-spokes전략은 교통분야에서 널리 이용되는 네트워크전략이지만, 철도의 경우 대개 이 전략을 이용하기 어려운 네트워크구조를 가지고 있어 그 적용사례를 찾아보기 어렵다. 그러나 유럽에서는 철도망이 도로망처럼 조밀하게 형성되어 있다는 점과 환적 처리기술의 발달로 90년대 초부터 이 전략이 철도화물운송부문에도 도입되기 시작했다. 이러한 관점에서 본 논문은 철도화물운송망에서의 hub-and-spokes전략을 구현하는 서비스네트워크 디자인모형을 개발하고, 모델의 실제철도망에의 적용성을 평가한다. 개발되는 모형이 전략모형임에도 불구하고 모형에서는 일반화된 운영비용 외에 열차속도, 서비스빈도, 터미널에서의 화물처리속도 등에 따른 시간지체비용도 고려되었다. 시간지체비용의 고려에 따라 야기되는 비선형 목적함수는 빈도별 서비스결정변수의 설정을 통해 선형화되어 결과적으로 모형은 선형 binary정수 최적화문제로 표현되었다. 규모가 큰 네트워크의 경우 해도출의 어려움 때문에 본 논문은 전체문제의 분할(decomposition)에 기초한 휴리스틱방법((heuristic method)으로 해결한다. 해도출의 효율성을 높이기 위해 서비스빈도개선과 관련하여 세 알고리즘이 개발되었다. 개발된 알고리즘은 유럽의 실제네트워크를 기초로 도출한 4개의 테스트문제에 적용되어, 해의 정확도와 해 도출의 효율성이 비교·평가되었다.