• Proceedings of the Korea Database Society Conference
• /
• 1999.06a
• /
• pp.415-421
• /
• 1999

Altman의 연구(1965, 1977)나 Beaver의 연구(1986)와 같은 전통적 예측모형은 분석자의 판단에 따른 예측도가 높은 재무비율을 선정하여 다변량판별분석(MDA： multiple discriminant analysis), 로지스틱회귀분석 등과 같은 통계기법을 주로 이용해 왔으나 1980년 후반부터 인공지능 기법인 귀납적 학습방법, 인공신경망모형, 유전모형 둥이 부실기업예측에 응용되기 시작했다. 최근 연구에서는 인공신경망을 활용한 변수 및 모형개발에 관한 보고가 있다. 그러나 지금까지의 연구가 주로 기업의 재무적 비율지표를 고려한 모형에 치중되었으며 정성적 자료인 비재무지표에 대한 검증과 선정이 자의적으로 이루어져온 경향이었다. 또한 너무 많은 입력변수를 사용할 경우 다중공선성 문제를 유발시킬 위험을 내포하고 있다. 본 연구에서는 부실기업예측모형을 수립하기 위하여 정량적 요인인 재무적 지표변수와 정성적요인인 비재무적 지표변수를 모두 고려하였다. 재무적 지표변수는 상관분석 및 요인분석들을 통하여 유의한 변수들을 도출하였으며 비재무적 지표변수는 조직생태학내에서의 조직군내 조직사멸과 관련된 생태적 과정에 대한 요인들 중 조직군 내적요인으로 조직의 연령, 조직의 규모, 조직의 산업밀도를 도출하여 4개의 실험집단으로 분류하여 비재무적 지표변수를 보완하였다. 인공신경망은 다층퍼셉트론(multi-layer perceptrons)과 역방향 학습(back-propagation )알고리듬으로 입력변수와 출력변수, 그리고 하나의 은닉층을 가지는 3층 퍼셉트론(three layer perceptron)을 사용하였으며 은닉충의 노드(node)수는 3개를 사용하였다. 입력변수로 안정성, 활동성, 수익성, 성장성을 나타내는 재무적 지표변수와 조직규모, 조직연령, 그 조직이 속한 산업의 밀도를 비재무적 지표변수로 산정하여 로지스틱회귀 분석과 인공신경망 기법으로 검증하였다. 로지스틱회귀분석 결과에서는 재무적 지표변수 모형의 전체적 예측적중률이 87.50%인 반면에 재무/비재무적 지표모형은 90.18%로서 비재무적 지표변수 사용에 대한 개선의 효과가 나타났다. 표본기업들을 훈련과 시험용으로 구분하여 분석한 결과는 전체적으로 재무/비재무적 지표를 고려한 인공신경망기법의 예측적중률이 높은 것으로 나타났다. 즉, 로지스틱회귀분석의 재무적 지표모형은 훈련, 시험용이 84.45%, 85.10%인 반면, 재무/비재무적 지표모형은 84.45%, 85.08%로서 거의 동일한 예측적중률을 가졌으나 인공신경망기법 분석에서는 재무적 지표모형이 92.23%, 85.10%인 반면, 재무/비재무적 지표모형에서는 91.12%, 88.06%로서 향상된 예측적 중률을 나타내었다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.03a
• /
• pp.415-421
• /
• 1999

Altman의 연구(1965, 1977)나 Beaver의 연구(1986)와 같은 전통적 예측모형은 분석자의 판단에 따른 예측도가 높은 재무비율을 선정하여 다변량판별분석(MDA:multiple discriminant analysis), 로지스틱회귀분석 등과 같은 통계기법을 주로 이용해 왔으나 1980년 후반부터 인공지능 기법인 귀납적 학습방법, 인공신경망모형, 유전모형 등이 부실기업예측에 응용되기 시작했다. 최근 연구에서는 인공신경망을 활용한 변수 및 모형개발에 관한 보고가 있다. 그러나 지금까지의 연구가 주로 기업의 재무적 비율지표를 고려한 모형에 치중되었으며 정성적 자료인 비재무지표에 대한 검증과 선정이 자의적으로 이루어져온 경향이었다. 또한 너무 많은 입력변수를 사용할 경우 다중공선성 문제를 유발시킬 위험을 내포하고 있다. 본 연구에서는 부실기업예측모형을 수립하기 위하여 정량적 요인인 재무적 지표변수와 정성적 요인인 비재무적 지표변수를 모두 고려하였다. 재무적 지표변수는 상관분석 및 요인분석들을 통하여 유의한 변수들을 도출하였으며 비재무적 지표변수는 조직생태학내에서의 조직군내 조직사멸과 관련된 생태적 과정에 대한 요인들 중 조직군 내적요인으로 조직의 연령, 조직의 규모, 조직의 산업밀도를 도출하여 4개의 실험집단으로 분류하여 비재무적 지표변수를 보완하였다. 인공신경망은 다층퍼셉트론(multi-layer perceptrons)과 역방향 학습(back-propagation)알고리듬으로 입력변수와 출력변수, 그리고 하나의 은닉층을 가지는 3층 퍼셉트론(three layer perceptron)을 사용하였으며 은닉층의 노드(node)수는 3개를 사용하였다. 입력변수로 안정성, 활동성, 수익성, 성장성을 나타내는 재무적 지표변수와 조직규모, 조직연령, 그 조직이 속한 산업의 밀도를 비재무적 지표변수로 산정하여 로지스틱회귀 분석과 인공신경망 기법으로 검증하였다. 로지스틱회귀분석 결과에서는 재무적 지표변수 모형의 전체적 예측적중률이 87.50%인 반면에 재무/비재무적 지표모형은 90.18%로서 비재무적 지표변수 사용에 대한 개선의 효과가 나타났다. 표본기업들을 훈련과 시험용으로 구분하여 분석한 결과는 전체적으로 재무/비재무적 지표를 고려한 인공신경망기법의 예측적중률이 높은 것으로 나타났다. 즉, 로지스틱회귀 분석의 재무적 지표모형은 훈련, 시험용이 84.45%, 85.10%인 반면, 재무/비재무적 지표모형은 84.45%, 85.08%로서 거의 동일한 예측적중률을 가졌으나 인공신경망기법 분석에서는 재무적 지표모형이 92.23%, 85.10%인 반면, 재무/비재무적 지표모형에서는 91.12%, 88.06%로서 향상된 예측적중률을 나타내었다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.1 no.2
• /
• pp.85-92
• /
• 2000

Warrants are often described as call potions written tv firms on their own stock. However, a call option is a pure side bet; i.e., none of the cash flows associated with the call's sale or exercise involves the firm. Issuing warrants on the other hand, can affect the firm's aggregate level of investment, composition of its capital structure. and the price of the stock on which warrant can be exercised. The problem of the warrant pricing can be solved by using of multivariate data analysis techniques, such as regression analysis or discriminant analysis, instead of OPM. The value of this approach is that we can evlauate the relative importance of each independent variable which affect a price of a warrant. This study empirically examines the Japanese warrant pricing by multiple regression analysis using a sample or 300 observations traded on Tokyo Stock Exchange during the periods between 1995 and 1996.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.1
• /
• pp.185-193
• /
• 2017

This study examined the family functions and satisfaction of life according to the types of family rituals and investigated the effects of family rituals on family function and satisfaction of life. The survey questionnaire was administered to 136 married women. SPSS Win program with MANOVA and Multiple Regression Analysis was used for data analysis. The main study findings were as follows. (1)The types of family rituals were significantly different according to age and satisfaction of life was significantly different according to the levels of family economy. (2)The influence of family ritual on family functions differed according to the subtype of family rituals. (3)The influence of family ritual on satisfaction of life differed according to the subtype of family rituals. This study indicated that it is necessary to develop affect, routine, and deliberateness for increasing family function and life satisfaction in married women.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.3
• /
• pp.149-156
• /
• 2019

This study investigated the effects of academic self-efficacy and resilience on perception of professionalism. The survey questionnaire was completed by 313 pre-service early childhood teachers. SPSS Win program was used to perform MANOVA and Multiple Regression Analysis. (1)The self-regulatory efficacy was significantly different according to grade and experience of child care practice. (2)The control and sociality, positivity were significantly different according to grade and experience of child care practice. (3)The perception of professionalism was significantly different according to grade and experience of child care practice. (4)The self-regulatory efficacy and the difficulty preference affected perception of professionalism. (5)The positivity and control affected perception of professionalism. This study indicated that it is necessary to make efforts for increasing perception of professionalism in pre-service early childhood teachers.

• Journal of the Korean Geotechnical Society
• /
• v.37 no.6
• /
• pp.29-37
• /
• 2021

Since many geotechnical structures are constructed on a weathered granite layer, it is important to evaluate their characteristics. As a seismic design is the more important nowadays, the demands to estimate a shear wave velocity (V_S) based on acceptable methods are increasing. In this study, an empirical equation predicting V_S of the weathered granite layer is suggested based on the nonlinear multiple variable regression analysis whose independent variables are both SPT (Standard penetration test)-N₆₀ and chemical weathering index. It is concluded that the accuracy of the empirical equation estimating V_S of the weathered granite layer increases when it considers the chemical weathering index as an additional independent variable compared to the result of simple regression analysis using only N₆₀.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.9
• /
• pp.2530-2534
• /
• 2009

This study was performed to evaluate relationship between health behaviors of local residents and stomach cancer incidence. Case-control study was carried out on 91 cases matched for age and sex with 91 controls in a general hospital in Seoul. It was surveyed using questionnaires and interview from December 20, 2008 to January 30, 2009. Major findings are as follows. First, significant negative association was observed in recreational activity ; OR for active group was 0.23, 0.41 relative to the inactive group. Secondly alcohol consumption was a significant risk factor; relative to non-drinker for drinker of more than 30 years(OR=5.17 95% Cl=1.78-19.95). Thirdly, Eating broiled food showed a positive significant association with risk for stomach cancer(OR=3.79 95% Cl=1.61-9.84) for consumption of three times more per week. As a result of multiple logistic regression analysis, income, BMI 10 years ago, recreational activity, alcohol consumption, eating broiled and salt food were associated with stomach cancer. In conclusion, the correlation of stomach cancer incidence can be found in recreational activity, alcohol consumption, eating broiled and salt food, income, and BMI 10 years ago.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.36 no.1
• /
• pp.12-20
• /
• 2020

Purpose: This study was designed to analyze the relationship between the presence and severity of depression and chewing problems (CPs) in a representative sample of the general population. Materials and Methods: Health surveys and examinations were conducted on a nationally representative sample (n = 8150) of Korean was conducted. CPs was determined by a simple survey response concerning "Do you feel uncomfortable about chewing your food because of problems with your mouth such as teeth, dentures and gums?" Depression was defined as individuals with a total score ≥ 10 on the Patient Health Questionnaire (PHQ)-9 survey. Data regarding demographics, socioeconomic history and comorbid health conditions were used to analyze adjusted logistic regression models. Results: In the Korean population, the prevalence of depression was significantly greater in individuals with CP (17.2%) than in those without CP (10.2%). On multivariate logistic regression analysis, the presence of depression was significantly associated with CPs (adjusted odd ratio [aOR]: 1.90, P < 0.001). The risk of CPs increased with increasing severity of depression as follows: severe depression (OR: 2.62, P < 0.001), moderately severe depression (OR: 2.19, P < 0.001). Conclusion: The presence of depression was significantly associated with CPs, especially in severely depressed individuals. Depression screening should be considered in treating CP patients.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.10
• /
• pp.4588-4597
• /
• 2012

This study aimed to investigate pain, perceived exercise barriers, and related factors in arthritis patients. The study was conducted on 140 arthritis patients (89 rheumatoid arthritis patients and 51 osteoarthritis patients) who visited C University Hospital in Daejeon due to arthritis and agreed to participate in a survey. The self-administered survey was conducted from February 2011 to September 2011. The results are as follow. The perceived exercise barriers increased in both groups as the education level and monthly household income decreased. In the rheumatoid arthritis group, the pain became more intense as the self-rated health status became poorer while the perceived exercise barriers became more significant as the self-rated health status became poorer and the patients had other diseases besides arthritis. In the osteoarthritis group, the pain was greater as the duration of arthritis was longer and the self-rated health status became poorer. In both groups, the perceived exercise barriers became more significant as the exercise stage approached the pre-planning stage. The results of the multiple linear regression analysis showed that the most influential factor on pain in both groups was the 'very bad' self-rated health status. In the rheumatoid arthritis group, influential factors on the perceived exercise barriers were the exercise stage and the education level while for the osteoarthritis group, they were the exercise stage and the monthly household income. In conclusion, the results of this study showed that the pain and perceived exercise barriers of arthritis patients were associated with various factors such as the education level, monthly household income, self-rated health status, and exercise stage, which suggests that these factors should be considered in the planning of exercise programs.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.8
• /
• pp.569-581
• /
• 2016

The purpose of this study was to define the quality of nursing services in relation to a nurse's psychosocial factors (type A behavior pattern, self-esteem, locus of control, and anxiety), job stress, and fatigue. This study targeted 503 nurses currently working at one of six general hospitals in Daejeon city, the data were collected by a self-administered questionnaire, which was surveyed from April 1 to June 30, 2014. According to multivariate logistic regression analysis, the high self-esteem, locus of control, and strong support by supervisor and co-workers decrease the risk of a low career quality. In contrast, high situational fatigue and overall fatigue increase the risk of a low career quality. According to the results of hierarchical multiple regression analysis, the quality of nursing services increased with age, the subjective health condition was satisfying, aptitude corresponded to the jobs, self-esteem was high, anxiety was low, support by co-workers was strong, and the overall fatigue was low. Above results suggest that the quality of nursing services of nurses who work at a general hospital has a strong correlation with the psychosocial factors, job stress, and fatigue.