• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38C no.2
• /
• pp.208-212
• /
• 2013

Due to the singularity of the within-class scatter, linear discriminant analysis (LDA) becomes ill-posed for small sample size (SSS) problems. An extension of LDA, the null space-based LDA (NLDA) provides good discriminant performances for SSS problems. In this paper, by applying the Lagrange technique, the procedure of transforming the problem of finding the feature extractor of NLDA into a linear equation problem is derived.

• Communications for Statistical Applications and Methods
• /
• v.22 no.5
• /
• pp.475-485
• /
• 2015

Similar to regression, many measures to detect influential data points in discriminant analysis have been developed. Many follow similar principles as the diagnostic measures used in linear regression in the context of discriminant analysis. Here we focus on the impact on the predicted classification posterior probability when a data point is omitted. The new method is intuitive and easily interpretable compared to existing methods. We also propose a graphical display to show the individual movement of the posterior probability of other data points when a specific data point is omitted. This enables the summaries to capture the overall pattern of the change.

• Health Policy and Management
• /
• v.5 no.1
• /
• pp.132-160
• /
• 1995

This study purported to acquire information necessary to improve the management of general hospitals. It tried to determine major indices which represent managerial performance of general hospitals and to identify the managerial characteristics of general hospitals which affect the major financial indices. Eighty-eight hospitals were chosen from 188 hospitals which were subject to standardization audit by the Korean Hospital Association. The results of a discriminant analysis are summarized as followings. First, when a single index was used to measure managerial performance of the sample hospitals, the ration of net profit to total capital was the best index and its discriminant power was 58.14%. The ratio of the number of boardmen((M. D.) and average daily medical cost were highly related to this index. Second, when two indices were used, income growth rte and the ration of net profit to total capital had the highest discriminant distinction ability. Their discriminant power was 61.9%. In this case, the ratio of the number of boardmen(M. D.) was significantly and highly related to the indices. Third, when all three indices-income growth rate, the ration of net profit to total capital and quick ratio - were used together, a discriminant function was statistically insignificant. Therefore, using all three indices was not useful in measuring managerial performance of the sample hospitals. In conclusion, using two indices-income growth rate and the ration of net profit to total capital-was better in measuring manegerial performance of general hospitals than using a single index. The independent variable which affected these indices was the ration of the number of boardmen. The discriminant function was : $D_{GI}=2.77+4.832\times(the ratio of the number of boardmen)$ *G=growth index(income growth rate) *I=profit index(the ration of net profit to total capital)

• Journal of the Korean Society of Physical Medicine
• /
• v.12 no.2
• /
• pp.9-20
• /
• 2017

PURPOSE: The purpose of this study is to understand the difference in the risk factors of coronary artery disease (CAD) between Korean and American CAD patients to determine the discriminant factor for each group, as well as to provide useful information to be reflected in the national concern of health. METHODS: Data were collected from 248 Korean and 107 American CAD patients who underwent either percutaneous coronary intervention or coronary artery bypass grafting. By using t-test and $X^2$-test, risk factors were compared between the Koreans and Americans. To elucidate which risk factor was the most discriminant for each group, logistic regression analysis was performed. RESULTS: All risk factors, except diastolic blood pressure, showed a significant difference between the two groups. $X^2$-test showed statistical significance with respect to the smoking rate between the female groups. Moreover, there was a statistically significant difference between the two groups regarding blood total cholesterol (TC) and triglyceride, and between the male groups, here was a statistically significant difference with respect to blood high-density lipoprotein cholesterol (HDL). Diabetes mellitus (DM) was the most discriminant factor for Korean patients while TC/HDL is the most discriminant for the Americans. CONCLUSION: The characteristics of CAD risk factors were determined to be different between Koreans and Americans in this study. TC/HDL was a discriminant factor for Americans while DM was a discriminant factor for Koreans. This result implies that DM should primarily be given attention to prevent CAD in Korean adults.

• Geomechanics and Engineering
• /
• v.13 no.5
• /
• pp.861-879
• /
• 2017

Coal and gas outburst is a serious dynamic disaster that occurs during coal mining and threatens the lives of coal miners. Currently, coal and gas outburst is commonly predicted using single indicator and its critical value. However, single indicator is unable to fully reflect all of the factors impacting outburst risk and has poor prediction accuracy. Therefore, a more accurate prediction method is necessary. In this work, we first analyzed on-site impacting factors and precursors of coal and gas outburst; then, we constructed a Fisher discriminant analysis (FDA) index system using the gas adsorption index of drilling cutting ${\Delta}h_2$, the drilling cutting weight S, the initial velocity of gas emission from borehole q, the thickness of soft coal h, and the maximum ratio of post-blasting gas emission peak to pre-blasting gas emission $B_{max}$; finally, we studied an FDA-based multiple indicators discriminant model of coal and gas outburst, and applied the discriminant model to predict coal and gas outburst. The results showed that the discriminant model has 100% prediction accuracy, even when some conventional indexes are lower than the warning criteria. The FDA method has a broad application prospects in coal and gas outburst prediction.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.25 no.1
• /
• pp.138-143
• /
• 2011

In spite of abundant clinical resources of stroke patients, the objective and logical data analyses or diagnostic systems were not established in oriental medicine. As a part of researches for standardization and objectification of differentiation of syndromes for stroke, in this present study, we tried to develop the statistical diagnostic tool discriminating the 4 subtypes of syndrome differentiation using the essential indices considering the sex. Discriminant analysis was carried out using clinical data collected from 1,448 stroke patients who was identically diagnosed for the syndrome differentiation subtypes diagnosed by two clinical experts with more than 3 year experiences. Empirical discriminant model(V) for different sex was constructed using 61 significant symptoms and sign indices selected by stepwise selection. We comparison. We make comparison a between discriminant model(V) and discriminant model(IV) using 33 significant symptoms and sign indices selected by stepwise selection. Development of statistical diagnostic tool discriminating 4 subtypes by sex : The discriminant model with the 24 significant indices in women and the 19 significant indices in men was developed for discriminating the 4 subtypes of syndrome differentiation including phlegm-dampness, qi-deficiency, yin-deficiency and fire-heat. Diagnostic accuracy and prediction rate of syndrome differentiation by sex : The overall diagnostic accuracy and prediction rate of 4 syndrome differentiation subtypes using 24 symptom and sign indices was 74.63%(403/540) and 68.46%(89/130) in women, 19 symptom and sign indices was 72.05%(446/619) and 70.44%(112/159) in men. These results are almost same as those of that the overall diagnostic accuracy(73.68%) and prediction rate(70.59%) are analyzed by the discriminant model(IV) using 33 symptom and sign indices selected by stepwise selection. Considering sex, the statistical discriminant model(V) with significant 24 symptom and sign indices in women and 19 symptom and sign indices in men, instead of 33 indices would be used in the field of oriental medicine contributing to the objectification of syndrome differentiation with parsimony rule.

• Women's Health Nursing
• /
• v.4 no.1
• /
• pp.123-138
• /
• 1998

The purpose of this study was to describe perimenstrual symptom severity levels and perimenstrual distress patterns of women. The study performed the discriminant analysis in which included seven factors : age, pariety, social support, menstrual socialization(mother's symptom, sister's symptom, and menstrual effect), attitude of sex role and depression. The subjects were 283 women that they were not pregnant or lactating, had at least one period in past three months, would understand the purpose of study and willingly accepted the participation. The data analysis was done by pc-SAS program after data collection from Nov. 20, 1997 to Dec. 18, 1997. The descriptive analysis was done to explore general characteristics of the subjects and the stepwise discriminant analysis was done to verify factors in relation to perimenstrual symptom severity levels(severe vs mild menstrual symptom group) and perimenstrual distress patterns(spasmodic vs congestive menstrual symptom group). The instruments were selected for this study from Interpersonal Support Evaluation List(ISEL) by Cohen and Hoberman(1983), Center for Epidemic Studies Depression(CES-D) by Radloff(1977), and Sex Role Attitude Scale by Yunok Suh(1995), Mother's symptom and sister's symptom measurements by Woods, Mitchell & Lentz(1995), and menstrual effect by Brooks-Gun & Ruble(1980). The major findings of this study are as follows : 1. Of the 283 women, 93 women(32.9%) were assessed to severe perimenstrual symptom group and 190 women(67.1%) were assessed to mild perimenstrual symptom group. Results from the stepwise discriminant analysis showed three factors, such as depression, menstrual effect, and age, significantly related to perimenstrual symptom severity and they explained 20% of the total variance. The linear discriminant equation included three factors related to perimenstrual symptom groups was showed(Z=1.445 depression+0.174 menstrual effect-0.054 age). The cutting score(Z) was 2.809. We classified the severe perimenstrual symptom group by more than the cutting score 2.809 and the mild perimenstrual symptom by less or equal than the cutting score 2.809. The correctedness of posterior probability from discriminant equation was 72% as two perimenstrual symptom group classifications. 2. Of the 264 women, 139 women(52.7%) were assessed to spasmodic perimenstrual distress group and women(47.3%) were assessed to congestive perimenstrual distress group. Results from the stepwise discriminant analysis showed two factors, such as depression, age, significantly related to perimenstrual distress groups and they explained 8% of the total variance. The linear discriminant equation included two factors related to perimenstrual distress group was showed(Z=-0.084 age-0.776 depression). The cutting score(Z) was -3.759. We classified the spasmodic perimenstrual distress group by more than cutting score -3.759 and the congestive perimenstrual distress group by less or equal than cutting score -3.759. The correctedness of posterior probability from discriminant equation was 65% as two perimenstrual distress group classifications.

• Journal of the Korean Statistical Society
• /
• v.30 no.2
• /
• pp.247-263
• /
• 2001

This paper concerns a problem of classification based on training dta. A framework of information geometry is given to elucidate the characteristics of discriminant functions including logistic discrimination and AdaBoost. We discuss a class of loss functions from a unified viewpoint.

• Communications for Statistical Applications and Methods
• /
• v.6 no.1
• /
• pp.43-52
• /
• 1999

The local influence method is adapted to quadratic discriminant analysis for the identification of influential observations affecting the estimation of probability density function probabilities and log odds. The method allows a simultaneous perturbation on all observations so that it can identify multiple influential observations. The proposed method is applied to a real data set and satisfactory result is obtained.

• Journal of the Korean Statistical Society
• /
• v.21 no.1
• /
• pp.1-13
• /
• 1992

This article deals with variable selection problem under a newly formed predictive heteroscedastic discriminant rule that accounts for mulitple homogeneous covariance matrices across the K multivariate normal populations. A general version of predictive discriminant rule, a variable selection criterion, and a criterion for stopping with further selection are suggested. In a simulation study the practical utilities of those considered are demonstrated.