• Asia pacific journal of information systems
• /
• v.19 no.2
• /
• pp.157-178
• /
• 2009

Corporate credit rating is a very important factor in the market for corporate debt. Information concerning corporate operations is often disseminated to market participants through the changes in credit ratings that are published by professional rating agencies, such as Standard and Poor's (S&P) and Moody's Investor Service. Since these agencies generally require a large fee for the service, and the periodically provided ratings sometimes do not reflect the default risk of the company at the time, it may be advantageous for bond-market participants to be able to classify credit ratings before the agencies actually publish them. As a result, it is very important for companies (especially, financial companies) to develop a proper model of credit rating. From a technical perspective, the credit rating constitutes a typical, multiclass, classification problem because rating agencies generally have ten or more categories of ratings. For example, S&P's ratings range from AAA for the highest-quality bonds to D for the lowest-quality bonds. The professional rating agencies emphasize the importance of analysts' subjective judgments in the determination of credit ratings. However, in practice, a mathematical model that uses the financial variables of companies plays an important role in determining credit ratings, since it is convenient to apply and cost efficient. These financial variables include the ratios that represent a company's leverage status, liquidity status, and profitability status. Several statistical and artificial intelligence (AI) techniques have been applied as tools for predicting credit ratings. Among them, artificial neural networks are most prevalent in the area of finance because of their broad applicability to many business problems and their preeminent ability to adapt. However, artificial neural networks also have many defects, including the difficulty in determining the values of the control parameters and the number of processing elements in the layer as well as the risk of over-fitting. Of late, because of their robustness and high accuracy, support vector machines (SVMs) have become popular as a solution for problems with generating accurate prediction. An SVM's solution may be globally optimal because SVMs seek to minimize structural risk. On the other hand, artificial neural network models may tend to find locally optimal solutions because they seek to minimize empirical risk. In addition, no parameters need to be tuned in SVMs, barring the upper bound for non-separable cases in linear SVMs. Since SVMs were originally devised for binary classification, however they are not intrinsically geared for multiclass classifications as in credit ratings. Thus, researchers have tried to extend the original SVM to multiclass classification. Hitherto, a variety of techniques to extend standard SVMs to multiclass SVMs (MSVMs) has been proposed in the literature Only a few types of MSVM are, however, tested using prior studies that apply MSVMs to credit ratings studies. In this study, we examined six different techniques of MSVMs: (1) One-Against-One, (2) One-Against-AIL (3) DAGSVM, (4) ECOC, (5) Method of Weston and Watkins, and (6) Method of Crammer and Singer. In addition, we examined the prediction accuracy of some modified version of conventional MSVM techniques. To find the most appropriate technique of MSVMs for corporate bond rating, we applied all the techniques of MSVMs to a real-world case of credit rating in Korea. The best application is in corporate bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. For our study the research data were collected from National Information and Credit Evaluation, Inc., a major bond-rating company in Korea. The data set is comprised of the bond-ratings for the year 2002 and various financial variables for 1,295 companies from the manufacturing industry in Korea. We compared the results of these techniques with one another, and with those of traditional methods for credit ratings, such as multiple discriminant analysis (MDA), multinomial logistic regression (MLOGIT), and artificial neural networks (ANNs). As a result, we found that DAGSVM with an ordered list was the best approach for the prediction of bond rating. In addition, we found that the modified version of ECOC approach can yield higher prediction accuracy for the cases showing clear patterns.

• Journal of agricultural medicine and community health
• /
• v.30 no.1
• /
• pp.101-111
• /
• 2005

Objectives: This study was to evaluate the compliance of hypertensive patients and its related factors registered at Community Health Practitioner Post(CHCP). Methods: 304 patients were interviewed by trained nursing students during one month(June~July 2004). The questionnaire included general charactristics, knowledge of hypertension, health education experience, constructs of Health Belief Model, self efficacy and so on. Compliance group was defined "having regularly medication and good life style". Good life style included regular exercise, non-smoking, little alcohol, low salt diet, weight control. Results: In compliance group 90.3% of man and 93.3% of woman were regularly taking hypertensive medicine, and 45.2% of man and 56.4% of woman were having good life style (compliance group). In both man and woman, the group of higher education were more compliance group, but were statistically significant were in man(p<0.05). In woman, the compliance group have significantly higher score in knowledge of hypertension(p(0.05). The compliance group have significantly higher self-efficacy score in both man and woman (p<0.05). In Health Belief Model, susceptibility and benefit were statistically significant in man, seriousness, benefit and barrier in woman(p<0.05). In multiple logistic regression analysis, education level and self efficacy in man and knowledge of hypertension, self-efficacy and benefit in woman were significant variables (p<0.05). Conclusions: It is very important to evaluate and modify life-style adding to having regularly medication in hypertensive patients registered at CHCP. To this, health education programs about benefit to compliance and the methods to improve self-efficacy should be developed for this patients.

• Radiation Oncology Journal
• /
• v.19 no.3
• /
• pp.224-229
• /
• 2001

Purpose : The aim of this study is to analyze the treatment failure patterns and the risk factors for locoregional or distant failure of uterine cervical carcinoma treated with radiation therapy. Materials and methods . A retrospective analysis was undertaken of 154 patients treated with curative radiation therapy in Gyeongsang National University Hospital from April 1989 through December 1997. According to FIGO classification, 12 patients were stage IB, 24 were IIA, 98 were IIB, 1 were IIIA, 17 were IIIB, 2 were IVA. Results : Overall treatment failure rate was $42.1\%$ (65/154), and that of complete responder was $31.5\%$ (41/130). Among 65 failures, 25 failed locoregionally, another 25 failed distantly, and 15 failed locoregionally and distantly. Multivariate analysis confirmed tumor size (>4 cm) as risk factor for locoregional failure, and tumor size (>4 cm), pelvic lymph node involvement as risk factors for distant failure. Conclusion : On the basis of results of our study and recent published data of prospective randomized study for locally advanced uterine cervical carcinoma, we concluded that uterine cervical carcinoma with size more than 4 cm or pelvic lymph node involvement should be treated with concurrent chemoradiation.

• Journal of Preventive Medicine and Public Health
• /
• v.28 no.4 s.51
• /
• pp.845-862
• /
• 1995

We followed prospectively some hospital-delivered mothers to identify characteristics of those not initiated breast-feeding and predictors of breast-feeding discontinuation in monthly telephone interviews. Recruits were composed of 482 mothers who delivered their babies at one university hospital and one OB/GYN clinic in September to November 1991. Breast-feeding discontinuation was defined as switch to 100% formula lasting more than one week regardless of solid foods. Average age of the study subjects was 27.3 years of age(standard deviation 3.2). Multiple logistic regression analysis indicated native place, occupation, method of delivery and method of feeding considered to be better for maternal health were statistically significant(p<0.1) between initiators and non-initiators of breast feeding. In starting cohort(N=242) of those initiated breast-feeding, that median of breast-feeding discontinuation were 5 months and 25th and 75th percentiles were 3 and 9 months respectively. In Cox's proportional hazard model, mothers with $10\sim13$ years of education were 2.63 times (95% confidence interval, CI $1.50\sim4.60$) more likely to discontinue than those with less than 9 years of education and those with more than 13 years of education were 3.55 time (95% CI $1.99\sim6.33$). Compared with house wife, mothers with part-time jobs were 1.99 times (95% CI $0.86\sim4.57$) more likely to discontinue and those with employed full-time were 1.55 times (95% CI $0.96\sim2.51$). These results suggest that the predictors of initiation and discontinuation of breast-feeding may be different and different target populations should be selected to promote initiation and to prevent discontinuation of breast-feeding according to the period after birth.

• Korean Journal of Soil Science and Fertilizer
• /
• v.39 no.6
• /
• pp.329-333
• /
• 2006

Most of volcanic ash soils in South Korea are distributed in Jeju province which is an island placed on southern part of Korea and has steep slope mountain area. There are many soils containing high contents of organic matter (OM) derived from volcanic ash in Jejudo, also. Therefore, irrigation and drainage in volcanic ash soil different with general soil which has low OM content have to be applied with another management way, but studies searching appropriate methods for them are set on insufficient situation because the area of volcanic ash soil in South Korea is only 1.3% (130,000ha). This study was conducted for analysis of soil water content and irrigation quantity appropriate for crops cultivated in volcanic ash soil with high OM content. Although soils with different soil color have the same soil texture, soil water characteristics curve by soil color showed the difference of water retention capability by OM content. But, this characteristics classified with soil color could be unified by scaling technique with similitude analysis method which get dimensionless water content using a present water content, a residual water content and saturated water content (or water content at 10kPa). A relation of gravimetric soil water content (GSWC) and dimensionless water content by the results showed a form of power function. The dimensionless water content (DWC) express a relative saturation degree of present water content. This was also expressed by van Genuchten model which describe the relation between relative saturation degrees and matric potentials. These results on soil water characteristics curve (SWCC) of volcanic ash soil will be the basic of irrigation plan in area having high organic contents into soil.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.4
• /
• pp.135-151
• /
• 2021

This research attempts to analyze the factors influencing the entrepreneurial intention of employees in manufacturing field. In particular, key factors of entrepreneurship and personal characteristics explain a significant association with the intention to start-up. And study whether R&D support from public enterprise adjusts intention to entrepreneurial Intention. This study conducted a online survey on 292 small and medium-sized enterprise manufacturing employees in May 2020. Using linear regression model and binary logistic model. The main study results are the following: First, among the key factors(innovativeness, proactiveness, risk-taking) of entrepreneurship, proactiveness hardly influenced the opportunity competency. Second, among the factors(risk-taking propensity, locus of control, tolerance for ambiguity) of personal characteristics, locus of control hardly influenced the opportunity competency. Third, opportunity competency(opportunity recognition and opportunity evaluation) had positive influence to entrepreneurial intention. Fourth, the study investigated the mediated effect of opportunity competency. The result showed that among the factors of entrepreneurship and personal characteristics, only two factors that are proactiveness and locus of control were not mediated by opportunity competency. and opportunity evaluation was acted as a mediator between proactiveness and entrepreneurial Intention, compared with opportunity recognition. Lastly, public enterprise's R&D supporting moderated the entrepreneurial intention). Based on the result, the study showed that first, the key factors of entrepreneurship except for proactiveness and personal characteristics(risk-taking propensity, locus of control, tolerance for ambiguity) except for locus of control affect the intention to start-up, repeatedly. This results are explained that employees have not started a business yet. Second, research on start-up suggests the need to analyze factors differentiated before and after the start-ups. Based on the results, entrepreneurship and personal characteristics show that study on the effects of start-up intentions should be carried out before and after the actual start-up takes place, and can be used as effective data in policies to promoting start-ups in manufacturing field.

• Journal of the Korean Society of Child Welfare
• /
• no.59
• /
• pp.209-234
• /
• 2017

This study viewed schools as a cause of students dropping out and posited that dropping out of high school would vary depending on the characteristics and influencing factors of the school from which students were dropping out. Therefore, focusing on schools, we longitudinally investigated the change patterns of school dropout across high schools in the country, and the types of changes in dropping out of high school. In addition, we predicted the general characteristics of schools according to the type of school students were dropping out from, looked at the changes in the major factors (i.e., school violence and school counseling) affecting school dropout, and reviewed schools' long-term efforts and outcomes in relation to school dropout. For this purpose, KERIS EDSS's "Secondary School Information Disclosure Data" were used. The final model included data collected five years20122016) from high schools across the country. The results were as follows. First, in order to examine the longitudinal change patterns of dropping out of high schools, a latent growth models analysis was conducted, and it revealed that, as time passed, the dropout rate decreased. Second, growth mixture modeling was used to explore types according to the change patterns of the school students were dropping out from. The results showed three types: the "remaining in school" type, the "gradually decreasing school dropout" type, and the "increasing school dropping out". Third, the multinomial logistic regression was conducted to predict the general characteristics of schools by type. The results showed that public schools, vocational schools, and schools with a large number of students who have below the basic levels in Korean, English and mathematics were more likely to belong to the "increasing school dropout" type. Further, the larger the total number of students, the higher the probability of belonging to the "remaining in school" type or the "gradually decreasing school dropout" type. Lastly, growth mixture modeling was used to analyze the trend of school violence and school counseling according to the three types. The focus was on the "gradually decreasing school dropout" type. In the case of the "gradually decreasing school dropout" type, it was found that as time passed, the number of school violence cases and the number of offenders gradually decreased. In addition, in terms of change in school counseling the results revealed that the number of placement of professional counselors in schools increased every year and peer counseling was continuously promoted, which may account for the "gradually decreasing school dropout" type.

• Korean Journal of Community Nutrition
• /
• v.23 no.6
• /
• pp.475-487
• /
• 2018

Objectives: This study examined the consumption of Han-sik and its association with the years of residence in Korea and the socioeconomic status among Filipino immigrant women of the Filipino Women's Diet and Health Study (FiLWHEL). Methods: A total of 474 Filipino women married to Korean men were included in the analysis. Their dietary intake was assessed using a single-day 24-hour recall. The participants provided information on the demographics, socioeconomic, and health-related factors through face-to-face interviews. The generalized linear model and logistic regression model were used to examine the association between the socioeconomic status and consumption of Han-sik. Results: The mean age of the participants was 34.3 years old, and the average duration of residence in Korea was 8.2 years. Among 474 Filipino women, a total of 467 consumed Han-sik, with an average of 6.8 food items per day. The Han-sik foods that the participants consumed most frequently were rice, cabbage kimchi, mixed-grain rice, and fried eggs. The average ratio of Han-sik was 58.57%. The ratio of Han-sik showed no significant associations with the years of residence, years of living together with their husband, education levels, total annual family income, or linguistic competence of Korean. However, the ratio of Han-sik use was associated with cohabitation with parents-in-law; the odds ratio (95% confidence interval) was 2.41 (1.18-4.92, p-trend = 0.002) comparing the fourth quartile with the first quartile of the Han-sik ratio. Conclusions: Filipino immigrant women in the FiLWHEL study consumed a larger number of Han-sik than Philippine foods. In addition, cohabitation with their parents-in-law was associated with the consumption of Han-sik. Further epidemiologic studies will be needed to determine how the diet affects the health and wellbeing of immigrant women in Korea.

• 한국노년학
• /
• v.37 no.1
• /
• pp.83-102
• /
• 2017

Expending on a life course perspective, this study explores the long-term and short-term reciprocity in parent-child relationships in Korean context. Since the reasons for providing filial support are believed to differ by gender, we focused on how a child's gender affects both types of reciprocity. Data were collected from middle-aged sons (N=726) and daughters (N=883) with at least one surviving parent. Logistic regression was then conducted in order to examine the relations between the support a child currently provides to parents and the current or previous support received from the parents. Dependent variables are financial and instrumental support that middle-aged child currently provide to the parents. The financial and instrumental support a child received from the parents within a year are included in the model as an independent variable to assess short-term reciprocity. The level of financial support a child has received during the transition to adulthood process is included in the model as a independent variable to explore long-term reciprocity. Result supports the existence of gender differences in the long-term reciprocity. Daughters provided instrumental support in response to the financial support that they had received from parents during the transition to adulthood process. However, for sons, this tendency was not found. When it comes to financial support, long-term reciprocity was observed neither for the sons nor for the daughters. Both sons and daughters are prone to provide financial support to the aged parents regardless of the level of financial support they had received during the transition to adulthood process. Short-term reciprocity was found both in sons and daughters. when they have been receiving a financial or an instrumental support from the aged parents within a year, they tend to provided instrumental support to the parents. This study shows that the aged parents still fulfill the reciprocal relationship to a certain degree. Secondly, we can conclude that the norm of reciprocity interplays with the norm of filial responsibility in contemporary Korea.

• 한국노년학
• /
• v.40 no.4
• /
• pp.707-727
• /
• 2020

This study aims to empirically analyze the role that perception of the probability of lonely death among the elderly people living alone plays in the relationship between perceived social support and preference for institutional care based on Andersen's expanded Behavioral Model (2002). The subjects (n=676) of this study were the elderly people living alone, extracted from the "2018 Seoul Aging Survey." With "perceived social support" as an independent variable, "preference for institutional care" as a dependent variable, and "perception of the probability of lonely death" as a mediator variable, we conducted a Binary Logistic Regression to follow the three steps of analyzing mediation effect, as suggested by Baron and Kenny (1986). The results showed that perceived social support has a negative effect on the preference for institutional care and perception of the probability of lonely death among the elderly people living alone; at the same time, perception of the probability of lonely death was found to have a positive effect on their preference for institutional care. Lastly, perception of the probability of lonely death was found to partially mediate the effect of perceived social support among the elderly people living alone in terms of their preference for institutional care. Based on these findings, the practical implications of this study can be summarized as follows. First, various programs and support should be provided to the elderly people living alone in order to enhance the level of perceived social support, a factor that has been confirmed to increase preference for institutional care among the elderly people living alone. Second, as the perception of the probability of lonely death was confirmed to be a psychosocial factor of the preference for institutional care, we need to promote education and support for older people living alone to prepare them for lonely death. These efforts are expected to form a foundations for implementing a community-based integrated care system, "Aging in Place," which is the policy direction required for older people care.