• The Korean Journal of Applied Statistics
• /
• v.19 no.3
• /
• pp.587-598
• /
• 2006

The number of variables exceeds the number of samples in microarray data. We propose a nonparametric local linear logistic classification procedure using orthogonal components for classifying high-dimensional microarray data. The proposed method is based on the local likelihood and can be applied to multi-class classification. We applied the local linear logistic classification method using PCA, PLS, and factor analysis components as new features to Leukemia data and colon data, and compare the performance of the proposed method with the conventional statistical classification procedures. The proposed method outperforms the conventional ones for each component, and PLS has shown best performance when it is embedded in the proposed method among the three orthogonal components.

• Korean journal of communication and information
• /
• v.49
• /
• pp.20-38
• /
• 2010

This paper was attempting to identify determinants of media repertoires based on new services and technologies. Using the regression and discriminant models, this study examined determinants that included five independent factors such as the degree of innovation, social networks, social influences, demographic variables and media uses. The analyses revealed that all of independent variables except the degree of innovation were significant determinants of media repertoires. Secondly, the results of discriminant analyses showed that terrestrial television use, age, disposable income were significant factors discriminating new media service adopters from the sample. For new media related technologies adopters, family income, and media uses such as newspaper, Internet and radio were significant discriminant variables.

• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2853-2864
• /
• 2018

Discriminant analysis predicts a class label of a new observation with an unknown label, using information from the existing labeled data. Hence, observed labels play a critical role in the analysis and we usually assume that these labels are correct. If the observed label contains an error, the data has label noise. Label noise can frequently occur in real data, which would affect classification performance. In order to resolve this, a comparative study was carried out using simulated data with label noise. In particular, we considered 4 different classification techniques such as LDA (linear discriminant analysis classifiers), QDA (quadratic discriminant analysis classifiers), KNN (k-nearest neighbour), and SVM (support vector machine). Then we evaluated each method via average accuracy using generated data from various scenarios. The effect of label noise was investigated through its occurrence rate and type (noise location). We confirmed that the label noise is a significant factor influencing the classification performance.

• Journal of Korean Society of Transportation
• /
• v.20 no.7
• /
• pp.143-153
• /
• 2002

본 연구는 교통사고의 발생원인에 대한 인식유형과 감소대책에 대한 인지 유형별 영향요인의 정도를 분석하기 위하여 수량화이론 II류와 CHAID 분석법을 이용하여 분류모델과 판별모델을 구축하였다. 수량화이론 II류에 의한 교통사고 발생원인에 대한 인식 유형별 영향요인 판별모델은 전체 적중률이 78.4%로 매우 높게 나타났다. 편상관계수는 설명변수의 항목 중 학력, 성별, 운전경력 년 수, 소유 차종의 순으로 영향을 미치고 외적 변수인 교통사고 발생원인에 대한 유형에서는 기여 정도가 교통단속 부재 > 교통체계 미비 > 승용차 과다 사용 >잘못된 의식 때문의 순으로 나타났다. 교통사고 감소 대책에 대한 인지유형별 영향요인 판별모델은 전체 적중률이 59.9%로 높게 나타났으며, 편상관 계수는 학력, 성별, 운전경력 연수, 연령의 순으로 영향을 미치고 있고, 외적 변수인 교통사고 감소 대책에 대한 유형에서는 기여 정도가 교통단속 강화 > 대중교통수단 이용 유도 > 교통체계 개선 > 의식 개혁의 순으로 나타났다. 또한 CHAID 분석법에 의한 교통사고 발생원인에 대한 인식 유형별 영향요인 분류모델에 있어서는 예측변수로 학력, 연령, 성별, 통행수단의 네 가지 변수가, 교통사고의 감소 대책에 대한인지 유형별 영향요인 분류모델에 있어서는 학력, 운전경력 연수, 성별 그리고 통행수단의 네 가지 변수가 카이제곱 통계량 이 5%의 유의수준에서 유의한 것으로 판단되었다. 교통사고 발생원인 인식과 감소 대책의 인지 유형에 대한 빈도분석과 교차분석은 의식과 관련한 유형이 가장 높게 나타났으나 판별.분류모델에서는 교통단속과 관련한 유형이 기여 정도가 높고 의식 관련 유형이 상대적으로 낮게 나타나는 등 반대양상을 보이고 있어 심리적으로 내재되어 있고 표면에 잘 드러나지 않았던 의식 수준의 낮음이 분류모델을 통해서 명확하게 드러났다.

• 한국사회복지학회:학술대회논문집
• /
• 2008.04a
• /
• pp.248-254
• /
• 2008

• Korean Journal of Adult Nursing
• /
• v.19 no.3
• /
• pp.389-400
• /
• 2007

Purposes: This study was designed to explore the stage distribution of subjects according to stage of change for calcium intake and for exercise, and to identify factors that could discriminate among subjects in various stages. Methods: The sample consisted of 142 subjects who had taken bone mineral densitometry tests. The instruments used in this study were the Stage Placement Instrument for Calcium Intake and Exercise, the Osteoporosis Health Belief Scale and the Osteoporosis Knowledge Test, and the Osteoporosis Self Efficacy Scale. Data were analyzed using chi square, ANOVA, and discriminant analysis by using the SPSS 12.0 program. Results: For calcium stages, economic level, calcium knowledge, positive social norms for calcium intake, & educational level showed high standardized canonical discriminant function coefficients. For exercise stages, exercise efficacy, susceptibility, exercise benefit, educational level, positive social norm to exercise, educational level, and exercise barrier showed high standardized canonical discriminant function coefficients. Conclusion: This study implies that bone mass promoting program incorporating a stages of change model can be applied as useful nursing intervention.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.435-449
• /
• 1998

In this article we have introduced a Bayes criterion for the variable selection in multiple discriminant analysis (MDA). The criterion is a default Bayes factor for the comparision of homo/heteroscadasticity of the multivariate normal means. The default Bayes factor is obtained from a development of the imaginary training sample method introduced by Spiegelhalter and Smith (1982). Based an the criterion, we also provided a test for additional discrimination in MDA. The advantage of the criterion is that it is not only applicable for the optimal subset selection method but for the stepwise method. More over, the criterion can be reduced to that for two-group discriminant analysis. Thus the criterion can be regarded as an unified alternative to variable selection criteria suggested by various sampling theory approaches. To illustrate the performance of the criterion, a numerical study has bean done via Monte Carlo experiment.

• Journal of Korea Technology Innovation Society
• /
• v.14 no.3
• /
• pp.664-688
• /
• 2011

This study aims to explore the variables which determine performance of inter-firm R&D cooperation. As the dependent variable is categorical - whether the new product developed by the inter-firm cooperation were sold or not-and the independent variables were interval, discriminant analysis was used. The independent variables were composed of degree of inter-firm cooperation, experience of cooperation, market attractiveness, R&D intensity, resources and competences of enterprise and efficiency of government support. A total of 144 responses were obtained. The results indicate that the degree or inter-firm cooperation is the best predictor of the performance, followed by market attractiveness, R&D intensity and resources/competences of enterprises. Whereas, the experience of cooperation and efficiency of government support program were not statistically significant predictors. The hit ratio or the percentage of cases correctly classified was 66.2%. We derived several implications of these findings in an effort to guide subsequent inquiry.

• Journal of the Korean Regional Science Association
• /
• v.34 no.2
• /
• pp.65-77
• /
• 2018

The purpose of this study is to compare and analyze factors in country risk between Cambodia and Vietnam. OECD and the Export-Import Ban of Korea assess country risk of Cambodia more highly than Vietnam. As results of the parametric tests for evaluation factors on the basis of country risk classification, the economic growth rate, the foreign trade index, and the foreign exchange reserves among the economic risks with the corruption index as the political and social risk have statistically significant effect on the difference between country risks of two countries. However, discriminant factor analysis indicates that the economic growth rate, the foreign exchange reserves, and the corruption index are key variables, which represent the difference between country risks of Cambodia and Vietnam. Consequently, the government of Cambodia needs to try to root out the corruption and to expand trade through increasing export for lowering the country risk to the level of Vietnam. Vietnam would also need to focus on attaining the sustainable high economic growth rate and increasing the foreign exchange reserves.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.2
• /
• pp.235-245
• /
• 2012

Logistic discrimination is an useful statistical technique for quantitative analysis of financial service industry. Especially it is not only easy to be implemented, but also has good classification rate. Generalized additive model is useful for credit scoring since it has the same advantages of logistic discrimination as well as accounting ability for the nonlinear effects of the explanatory variables. It may, however, need too many additive terms in the model when the number of explanatory variables is very large and there may exist dependencies among the variables. Mixtures of factor analyzers can be used for dimension reduction of high-dimensional feature. This study proposes to use the low-dimensional factor scores of mixtures of factor analyzers as the new features in the generalized additive model. Its application is demonstrated in the classification of some real credit scoring data. The comparison of correct classification rates of competing techniques shows the superiority of the generalized additive model using factor scores.