• Communications for Statistical Applications and Methods
• /
• v.30 no.5
• /
• pp.437-451
• /
• 2023

Generalized linear models and generalized linear mixed models (GLMMs) are fundamental tools for predictive analyses. In insurance, GLMMs are particularly important, because they provide not only a tool for prediction but also a theoretical justification for setting premiums. Although thousands of resources are available for introducing GLMMs as a classical and fundamental tool in statistical analysis, few resources seem to be available for the insurance industry. This study targets insurance professionals already familiar with basic actuarial mathematics and explains GLMMs and their linkage with classical actuarial pricing tools, such as the Buhlmann premium method. Focus of the study is mainly on the modeling aspect of GLMMs and their application to pricing, while avoiding technical issues related to statistical estimation, which can be automatically handled by most statistical software.

• Journal of the Korean Statistical Society
• /
• v.30 no.2
• /
• pp.219-230
• /
• 2001

How do you carry out data analysis\ulcorner There are few texts and little theory. One approach could be to use a pattern language, an idea which has been successful in field as diverse as town planning and software engineering. Patterns for data analysis are defined and discussed, illustrated with examples.

• The Korean Journal of Malacology
• /
• v.28 no.4
• /
• pp.363-375
• /
• 2012

The gametogenic cycle, the number of spawning seasons per year and first sexual maturiity of the pen shell, Atrina (Servatrina) pectinata, were investigated by quantitative statistical analysis using an Image Analyzer System. Compared two previous results (the spawning periods in the reproductive cycles in 1998 and 2006) by qualitative histological analysis with the present results by quantitative statistical analysis, there are some differences in the spawning periods: the spawning period (June to September) by quantitative statistical analysis was one month longer than those of two previous reports (June to July or June to August) by qualitative histological analysis. However, the number of spawning seasons studied by the qualitative and quatitative analyses occurred once per year. In quantitative statistical analysis using an image analyzer system, the patterns of monthly changes in the percent (%) of the areas occupied by follicles to the ovary area in females (or that of the areas occupied by spermatogenic stages to the testis area in males) showed a maximum in May, and then sharply droped from June to September, 2006. From these data, it is apparent that the spawning season of A. (S.) pectinata occurred once a year from June to September, indicating a unimodal gametogenic cycle during the year. Shell heights of sexually mature pen shells (size at 50% of group sexual maturity, $GM_{50}$) that were fitted to an exponential equation were 15.81 cm in females and 15.72 cm in males (considered to be one year old).

• Journal of Periodontal and Implant Science
• /
• v.30 no.3
• /
• pp.699-707
• /
• 2000

The purpose of this study was to analyze the statistical errors of articles in the Journal of Korean Academy of Periodontology from 1973 to 1999. Of the 662 articles examined, 263 were included which analyzed the data. They were classified into 2 groups with time lapse; group 1: 1973∼1989, group 2: 1990∼1999. Authors made checklists for analyzing the data and detecting the errors and analyzed them with professional statistician. The results were as follows: 1. Of 263 atricles which applied statistical method, 40(19.3%) was in group 1, 223(49.0%) in group2. 2. In the number of statistical method applied, 170(64.6%) were analyzed with 1 statistical method, 73(27.8%) with 2 methods, 18(6.8%) with 3 methods, and 2(0.8%) with 4 methods 3. The number of statistical methods applied was 14, and they were applied in order of 119 of ANOVA, 72 of Student t-test, 63 of Paired t-test, 36 of CORRELATION, and 21 of Mann- Whitney U test. 4. In 87(33.1%) of 263 articles and in 18 error items, statistical errors were found out. In group I, 9 items (55%) of error were found out, and were in order of 5 of Student t-test instead of Paired t-test, and 4 of unnecessary statistical analysis. In group II, 16 items (29.1%) of error were found out, and were in order of 22 of Student t-test instead of Paired t-test, 7 of no multiple comparison test after ANOVA, 6 of Student t-test instead of ANOVA, 6 of unnecessary statistical analysis, and 5 of ANOVA instead of Paired t-test. In conclusion, the results noted that statistical analyses were increased, but statistical errors were decreased with time. But authors suggest that researchers should refer to standard statistical texts and seek advice from professional statisticians to avoid the statistical errors.

• Environmental Analysis Health and Toxicology
• /
• v.30 no.sup
• /
• pp.5.1-5.5
• /
• 2015

Objectives To protect public health from risk, the Minister of Environment in Korea legislated an act concerning the registration and evaluation of chemical substances. In this study, we estimated the value of a statistical life (VSL) of carcinogenic chemicals to evaluate the socioeconomic analysis in Korea. Methods The estimation of the health benefit can be calculated through an individual's VSL and willingness to pay (WTP). To estimate the VSL and WTP, we used a contingent valuation method through a web-based survey. Results The survey is conducted with 1434 people living in Seoul and six large cities. An analysis of the survey is essential to review the distribution of the characteristics of the target population. The statistically significant variables affecting the WTP are location, age, household income, quality of life. Through the review of data, we secured statistical validity. The WTP was estimated as 41205 Korean won (KRW)/person, and the estimated VSL appeared as 796 million KRW/person. Conclusions There is a case in which the amount of statistical life value is estimated in connection with domestic environmental policy, fine dust, etc. However, there are no cases of evaluation for chemical. The utilization of this result is possible for conducting other study with chemicals.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.5
• /
• pp.645-652
• /
• 2016

The statistical modeling of input random variables is necessary in reliability analysis, reliability-based design optimization, and statistical validation and calibration of analysis models of mechanical systems. In statistical modeling methods, there are the Akaike Information Criterion (AIC), AIC correction (AICc), Bayesian Information Criterion, Maximum Likelihood Estimation (MLE), and Bayesian method. Those methods basically select the best fitted distribution among candidate models by calculating their likelihood function values from a given data set. The number of data or parameters in some methods are considered to identify the distribution types. On the other hand, the engineers in a real field have difficulties in selecting the statistical modeling method to obtain a statistical model of the experimental data because of a lack of knowledge of those methods. In this study, commonly used statistical modeling methods were compared using statistical simulation tests. Their advantages and disadvantages were then analyzed. In the simulation tests, various types of distribution were assumed as populations and the samples were generated randomly from them with different sample sizes. Real engineering data were used to verify each statistical modeling method.

• Proceedings of the KIEE Conference
• /
• 2002.07a
• /
• pp.67-69
• /
• 2002

This paper presents the statistical analysis of the circuit breaker's making current in large sacle industrical loads using EMTDC. Typically, the making current includes DC components as the phase angel of making instances varies and represents as a instantaneous valus not rms. Consequently, in this paper, the statistical analysis results of making current and steady state fault current for typical industrial loads presents.

• Journal of the Korean Statistical Society
• /
• v.29 no.2
• /
• pp.137-153
• /
• 2000

Conjoint analysis is a widely-used statistical technique for measuring relative importance that individual place on the product's attributes. Despsite its practical importance, the complexity of conjoint model makes it difficult to analyze. In this paper, w consider a Bayesian approach using Jeffrey's noninformative prior. We derive Jeffrey's prior and give a sufficient condition under which the posterior derived from the Jeffrey's prior is paper.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.72-76
• /
• 1990

A new statistical linearization technique for nonlinear system called covariance matching method is proposed in this paper. The covariance matching method makes the mean and variance of an approximated output be identical real functional output, and the distribution of the approximated output have identical shape with a given random input. Also, the covariance matching method can be easily implemented for statistical analysis of nonlinear systems with a combination of linear system covariance analysis.

• Research in Mathematical Education
• /
• v.14 no.2
• /
• pp.143-172
• /
• 2010

In this paper, we present an analysis of how students understand some statistical terms, mainly from inferential statistics, which are taught at the high school level. We focus our analysis on those terms that present more difficulties and are persistent in spite of having been studied until the college level. This analysis leads us to a hierarchical classification of responses at different levels of understanding using the SOLO theoretical framework.