Note on Use of R for No-intercept Model

도종두,김태윤 한국데이터정보과학회 2006 한국데이터정보과학회지 Vol.17 No.2

There have been some controversies on the use of the coefficient of determination for linear no-intercept model. One definition of the coefficient of determination, , is being widely accepted only for linear no-intercept models though Kvalseth (1985) demonstrated some possible pitfalls in using such . Main objective of this note is to report that is not a desirable measure of fit for the no-intercept linear model. In fact it is found that mean square error(MSE) could replace efficiently in most cases where selection of no-intercept model is at issue.

A study on development of economic instability index

도종두,송규문,김태윤 한국데이터정보과학회 2004 한국데이터정보과학회지 Vol.15 No.2

Kim et al.. (2003) developed an Economic Instability Index (EII) by using mean squared error (MSE) from the neural network (NN) trained on the 1995 KOSPI. In this paper we study validity of the NN. For this we compare the NN with the well known Box-Jenkins linear auto-regressive processes. Our conclusive understanding of the problem is that the NN provides quite effective EII because it tends to overfit.

회상요법이 치매노인의 인지기능과 기억 및 문제행동, ADL 증진에 미치는 영향

전은석,도종두 한국보건복지학회 2004 보건과 복지 Vol.7 No.-

Nowadays senile dementia is on the rise as consequence of serious social problem because of it is old aged society by improvement of medical service. However remedial approach is not being done because of we still don't have an infallible cure of dementia and dementia is been hard to cure. Thus it is introduced in this study to the concept of Reminiscence therapy, which is one of the remedies. I tried to find out which remedy is the best to cure demented elderly and which is been able to use on field. For this study, I applied a developed Reminiscence therapy program in actual practice. First, I chose nine demented elderly who are lived in 'Sunwoo Dementia Consulting Center' located in Daejon metropolitan city, and I separated them in two groups for this study. Second, I applied Reminiscence therapy to first group, and I did not apply any Reminiscence therapy to second group. The first group was applied the program, one time a week for fifty minutes, for about three months. The measurement of Cognitive function was practiced four times - one time before applying the Reminiscence therapy by HDS-R and ADAS-Cognitive, and one time during applying the program, and one time after ending of the program and one time after one month from ending the program. The measurement of Memory and Behavior problems function was practiced four times - one time before applying the Reminiscence therapy by MMSE, and one time during applying the program, and one time after ending of the program and one time after one month from ending the program. The measurement of ADL was practiced two times - one time before applying the Reminiscence therapy by the ADL, and one time after one month from ending the program. Effectiveness of Reminiscence therapy and possibility of application of Reminiscence therapy program are proved because of this study shows the Reminiscence therapy keeps well the Cognitive function, Memory and Behavior problems function and the ADL function of demented elderly, improves those functions, affects positively to the whole life. However we must develop Reminiscence therapy, which is proper to a trait of character, and expanded application is needed. In addition, temporary remedy must not be performed but continuous cure. This study will be complement by studying about special situation researcher cannot control something like unexpected death of demented elderly by other diseases.

사후검증(Back-testing)을 통한 다변량-GARCH 모형의 평가: 사례분석

황선영,최문선,도종두,Hwang, S.Y.,Choi, M.S.,Do, J.D. 한국통계학회 2009 응용통계연구 Vol.22 No.2

주식 수익률, 환율 등과 같은 금융 자료를 이해하는데 있어서 최근의 국제 금융위기를 통해 더욱 중요해진 이슈는 바로 변동성(volatility)이다. 변동성(조건부 이분산성)에 대한 모형은 Engle (1982)의 ARCH 모형과 Bollerslev (1986)의 GARCH 모형을 시작으로 수만은 연구가 이루어졌으며 특히 금융 시계열 분석에서는 시계열 자료들 간의 변동성을 함께 모형화 하는 MGARCH(multivariate GARCH) 모형이 널리 이용되고 있다. 추정된 MGARCH 모형들은 그 자체로서 여러 개의 변동성들 간의 시간에 따른 동적인 관계를 설명해주는 데 유용할 뿐만 아니라 추정된 (조건부)상관계수들은 hedge ratio 계산 또는 VaR 계산 등과 같이 금융시장에 대한분석에도 이용되고 있다. 본 논문에서는 국내 14개 최신 주가자료에 대한 MGARCH 분석을 수행하고 연관된 사후검증(back-testing)을 통해 MGARCH 모형들을 평가하고 있으며 사후검증 수치를 얻기 위한 S-PLUS 프로그램을 수록하였다. Current financial crisis triggered by shaky U.S. banking system adds to the emphasis on the importance of the volatility in controlling and understanding financial time series data. The ARCH and GARCH models have been useful in analyzing economic time series volatilities. In particular, multivariate GARCH(MGARCH, for short) provides both volatilities and conditional correlations between several time series and these are in turn applied to computations of hedge-ratio and VaR. In this short article, we try to assess various MGARCH models with respect to the back-testing performances in VaR study. To this end, 14 korean stock prices are analyzed and it is found that MGARCH outperforms rolling window, and BEKK and CCC are relatively conservative in back-testing performance.

데이터마이닝과 마케팅리서치에서 토빗모형의 응용에 관한 사례 연구

강현철,최호식,한상태,도종두,임민아 한국자료분석학회 2009 Journal of the Korean Data Analysis Society Vol.11 No.1

구조적으로 반응변수의 값이 제한된 범위(limited range) 안에서만 측정되는 자료를 빈번히 관찰할 수 있다. 이러한 경우 회귀분석 방법론에서 널리 사용되는 최소제곱추정법의 적용은 일반적으로 적절하지 않다고 알려져 있다. 본 연구에서는 반응변수가 제한된 범위 안에서 측정된 자료에 대하여 토빗모형(tobit model)과 선형회귀모형의 결과를 비교함으로써, 통상적인 최소제곱추정법이 실제 독립변수의 영향을 과소평가할 수 있음을 실증적으로 보이고자 한다. 특히 반응변수의 값이 하한 혹은 상한에 많이 집중되어 있는 경우 최소제곱추정법은 편의를 가진 또는 불일치하는 추정치를 제공할 수 있으며, 존재할 수 없는 반응변수의 예측값을 산출할 수도 있음을 살펴보았다. 또한 데이터마이닝과 마케팅리서치에서 전형적으로 얻어지는 실제 사례자료에 대하여 토빗모형을 적용함으로써, 토빗모형이 통상적인 회귀모형에 비하여 실제의 응용적, 해석적 측면에서 보다 효율적임을 보이고자 한다. It is not unusual for the response variable in a regression model to be subject to censoring or truncation. Tobit regression models are specific examples of such a situation, where for some observations the observed response is not the actual response, but the censoring value (often zero), and an indicator that censoring (from below) has occurred. In this case, applying the OLS method is become known the regression analysis methodological that is not appropriately. It is desirable to apply to the tobit model which reflects the structure of the limited dependent variable from a viewpoint of goodness of fit and construction. In this paper, applying the tobit model to real data in the fields of data mining and marketing research, we compare the result of the regression model with the tobit model to reflect the structure of the limited dependent variable.

보험사의 고객 이탈에 대한 예측모형 개발

한상태,강현철,최호식,도종두,신선화 한국자료분석학회 2009 Journal of the Korean Data Analysis Society Vol.11 No.1

Recently, with development of information technology, a great deal of data accumulate in companies' operational database. To manage these data more efficiently, many companies already set database of information such as Data Warehouse (DW) and they are trying to make good use of these data. In this study, we're going to introduce the real project format for utilizing these data at work. Especially we aim to show that how data mining is being used for searching information. Also, specifically, we are going to show you whole process of searching information modeling by using customers attribute and information of their transactions in insurance company. 최근 정보기술 분야의 급속한 발전과 더불어 기업들의 운영계 데이터베이스에는 엄청난 양의 데이터가 쌓이고 있다. 이러한 데이터를 보다 효율적으로 관리하기 위하여 많은 기업들에서는 데이터웨어하우스(DW)로 대변되는 정보계 데이터베이스를 이미 구축하였고 이를 활용하려는 움직임이 활발하게 진행되고 있다. 본 연구에서는 이러한 기업의 데이터를 업무에 활용하기 위한 실제 프로젝트의 모습을 소개하고자 한다. 특히 데이터마이닝이 지식발견에 어떻게 활용되고 있는가를 보여주고자 하는 것이 목적인데, 구체적으로 보험사의 고객 속성 및 거래 정보를 이용하여 지식발견 모델링의 구체적 프로세스를 보여줄 것이다.

여성노인에 대한 미용교육 프로그램의 효과

김보경 ( Bo Kyoung Kim ),박천만 ( Chun Man Park ),도종두 ( Jong Doo Do ) 한국미용학회 2008 한국미용학회지 Vol.14 No.2

This study focuses on female senior citizens at a general welfare center and aims to identify a beauty program`s effect on their depression, self-respect and life satisfaction. The study`s subjects were 90 female senior citizens who understood the study`s aims and purposes and agreed to take part in a beauty program. They were selected from the female senior citizens visiting the S. General Welfare Center located in the city of B. In consideration of their age, the selected subjects were equally divided into a control group and an experimental group. A beauty program was applied to the experimental group for eight weeks from May 5, 2007 to June 30, 2007. It consist of hand care, scalp and hair care, skin care, hair setting, make-up, hair dyeing, beauty health physical exercise and foot care. The duration of the class was 60 minutes per session, and was carried out once per week. An analysis was then conducted of the 45 subjects in the control group and 43 subjects in the experimental group, excluding two absentees. The results of the study are summarized below: 1. The experimental group showed a significant reduction in depression (p<0.01) after the program. 2. The experimental group significantly improved their self-respect, compared to the control group(p<0.05). 3. The control group`s life satisfaction declined on the other hand the experimental group`s life satisfaction improved significantly(p<0.01). Overall, the eight-week beauty program for female senior citizens effectively improved their quality of life by improving their depression, self-respect and life satisfaction. Further studies should develop programs that consider female senior citizens` features and needs.