• Journal of the Korean Data Analysis Society
• /
• v.20 no.6
• /
• pp.2733-2746
• /
• 2018

The two inverse regression estimation methods, SIR and SAVE to estimate the central space are computationally easy and are widely used. However, SIR and SAVE may have poor performance in finite samples and need strong assumptions (linearity and/or constant covariance conditions) on predictors. The two non-parametric estimation methods, MAVE and dMAVE have much better performance for finite samples than SIR and SAVE. MAVE and dMAVE need no strong requirements on predictors or on the response variable. MAVE is focused on estimating the central mean subspace, but dMAVE is to estimate the central space. This paper explores and compares four methods to explain the dimension reduction. Each algorithm of these four methods is reviewed. Empirical study for simulated data shows that MAVE and dMAVE has relatively better performance than SIR and SAVE, regardless of not only different models but also different distributional assumptions of predictors. However, real data example with the binary response demonstrates that SAVE is better than other methods.

• East Asian mathematical journal
• /
• v.37 no.5
• /
• pp.553-566
• /
• 2021

In the over-the-counter market, option's buyers could have a problem for default risk caused by option's writers. In addition, many participants try to maximize their benefits obviously in investing the financial derivatives. Taking all these circumstances into consideration, we deal with the vulnerable power options under a constant elasticity variance (CEV) model. We derive an analytic pricing formula for the vulnerable power option by using the asymptotic analysis, and then we verify that the analytic formula can be obtained accurately by comparing our solution with Monte-Carlo price. Finally, we examine the effect of CEV on the option price based on the derived solution.

• The Korean Journal of Rehabilitation Nursing
• /
• v.2 no.1
• /
• pp.29-44
• /
• 1999

This study was carried out to obtain basic data required to plan and develop Rehabilitation Day Care Program for the stroke Survivors at home in Korea. The subjects comprised of 118 stroke survivors who discharged from 4 hospitals in Seoul during the past 2 years. The data were collected from August 3, 1998 to September 18, 1998, through interviews with questionnaires about general characteristics, activities of dally living, depression and service need of rehabilitation day care program at the outpatient clinics by trained nursing graduates. Data were analyzed with descriptive analysis, Pearson's correlation analysis, and Stepwise multiple linear regression analysis using SPSS/WIN program. The results obtained are as follows ; 1. The mean score of the general need of rehabilitation day care program of stroke survivors was 2.78(range 1-4). The highest need among the service categories of the rehabilitation day care program was self-care and restorative activities category, and health services referral category, recreation category, psychosocial activities category in order. The needs of each category are as follows ; 1) In the health services referral category, the need for speech therapy was highest, followed by the need for physical therapy and occupational therapy. 2) In the psychosocial activities category, the need for self-help group was highest. 3) In the self-care and restorative activities category, the need for bathing was highest, followed by bowel training, and ambulation training. 4) The need for the recreation category was 2.62. 2. Among the need for the effect related to the utilization of day care program, the need for survivors' physical and psychological well-being was highest and was followed by the need for caregiver's physical and psychological wellbeing. Pearson's correlation analysis revealed following results ; 1. The need for rehabilitation day care program service displayed a correlation with the level of education, ADL, and the level of depression, and a reverse correlation with age. 2. The need for the effect related to the utilization of rehabilitation day program displayed a correlation with the level of education, ADL, and the level of depression. The stepwise multiple linear regression analysis revealed following results : 1. For the need for rehabilitation day care program service, 28.4% of the variance was initially explained by one variable, level of depression. The level of depression plus two variables, survivors' age and ADL, explained 34.2% of the variance in the need for rehabilitation day care program service. 2. For the need for the effect related to the utilization of rehabilitation day care program, 12.4% of the variance was initially explained by one variable, level of depression. The level of depression plus one variable, level of education, explained 20.4% of the variance in the need for the effect related to the utilization of rehabilitation day care program. In conclusion, above characteristics should be considered when we are planning to develop stroke survivors' rehabilitation day care program.

• Journal of Ginseng Research
• /
• v.35 no.1
• /
• pp.31-38
• /
• 2011

In order to distinguish the cultivation area of Panax ginseng, principal component analysis (PCA) using quantitative and qualitative data acquired from HPLC was carried out. A new HPLC method coupled with evaporative light scattering detection (HPLC-ELSD) was developed for the simultaneous quantification of ten major ginsenosides, namely $Rh_1$, $Rg_2$, $Rg_3$, $Rg_1$, Rf, Re, Rd, $Rb_2$, Rc, and $Rb_1$ in the root of P. ginseng C. A. Meyer. Simultaneous separations of these ten ginsenosides were achieved on a carbohydrate analytical column. The mobile phase consisted of acetonitrile-water-isopropanol, and acetonitrile-water-isopropanol using a gradient elution. Distinct differences in qualitative and quantitative characteristics for ginsenosides were found between the ginseng roots produced in two different Korean cultivation areas, Ganghwa and Punggi. The ginsenoside profiles obtained via HPLC analysis were subjected to PCA. PCA score plots using two principal components (PCs) showed good separation for the ginseng roots cultivated in Ganghwa and Punggi. PC1 influenced the separation, capturing 43.6% of the variance, while PC2 affected differentiation, explaining 18.0% of the variance. The highest contribution components were ginsenoside $Rg_3$ for PC1 and ginsenoside Rf for PC2. Particularly, the PCA score plot for the small ginseng roots of six-year old, each of which was light than 147 g fresh weight, showed more distinct discrimination. PC1 influenced the separation between different sample sets, capturing 51.8% of the variance, while PC2 affected differentiation, also explaining 28.0% of the variance. The highest contribution component was ginsenoside Rf for PC1 and ginsenoside $Rg_2$ for PC2. In conclusion, the HPLC-ELSD method using a carbohydrate column allowed for the simultaneous quantification of ten major ginsenosides, and PCA analysis of the ginsenoside peaks shown on the HPLC chromatogram would be a very acceptable strategy for discrimination of the cultivation area of ginseng roots.

• Journal of radiological science and technology
• /
• v.27 no.4
• /
• pp.67-74
• /
• 2004

In order to assess policies for training of radiological technologists and public health, this study was conducted from the mid July to the end of August 2003 to the subjects of 890 radiological technologists who work at 44 general hospitals in 16 cities and provinces of the country. For this purpose, an analysis of variance was performed by regions divided in Capital city, metropolitan areas and small- and medium-sized cities as well as by length of service. The outcomes look as follows : 1. From the analysis of replies to questions about the government health policy for radiological technologists, high scores were given to the items: "the Law of Medical Engineers should be revised from the pattern of supervision to assignment" and "the medical institutions should be legally obliged to secure certain number of radiological technologists". 2. From the analysis of variance by regions and groups, there was difference by regions in the items: "demand and supply of radiological technologists should be controlled" and "the license number of radiological technologists should be indicated in the medical insurance bill"(p<0.000). 3. From the analysis of variance as to training of radiological technologists, a significant difference was found by length of service in the items: "it is necessary to open a 4-year department of radiology and to revise curriculum" and "the legal system to indicate license number of radiological technologists in the medical insurance bill should be introduced". 4. The analysis of variance by length of service revealed that those who have served for more than 20 years favored "it is necessary to open a 4-year department of radiology and to revise curriculum" higher than those who have served for less than 5 years, showing a tendency the longer they work, the stronger they feel necessity".

• Korea journal of population studies
• /
• v.28 no.2
• /
• pp.131-164
• /
• 2005

The central focus of this research paper is to explore the recent structural core empirical unemployment reason of IT new professional manpower who graduated in higher education level in Korea. The final goal of this issue is to estimate and analyze rambling impediment in IT manpower labor market between 26 detail occupations and between detail occupations and 3 educational levels of new graduates and entrants who entered in IT 6 occupational groups composing 26 particular occupations. This study uses the 2004 data published by the Ministry of ICT and relies upon both one-way analysis of variance and two-way analysis variance to tackle the issue. According to the ANOVA results, 2 main points were delineated In the one-way analysis of variance method. the total sum of squares (SST) is decomposed into two major components: within sum of squares (SSW) (6 detail occupations, 85%) and between sum of squares (SSG) (26 detail occupation, 15%) The other major finding is that total sum of squares (SST) is decomposed into three major components: between detail occupations (SSG) (19.3%), between educations (SSB) and errors (SSE) (56%) in IT SIW development design group.

• Restorative Dentistry and Endodontics
• /
• v.39 no.1
• /
• pp.74-77
• /
• 2014

• Restorative Dentistry and Endodontics
• /
• v.39 no.2
• /
• pp.143-147
• /
• 2014

• Communications for Statistical Applications and Methods
• /
• v.5 no.2
• /
• pp.417-429
• /
• 1998

This paper compares powers of the two nonparametric tests under a variety of population distributions through a simulation study. Both tests require that the two underlying populations have the same variance, but this assumption is relaxed in some of the comparisons.

• Proceedings of the Korean Society of Environment and Ecology Conference
• /
• 2003.04a
• /
• pp.27-31
• /
• 2003