• Communications for Statistical Applications and Methods
• /
• 제11권2호
• /
• pp.297-311
• /
• 2004

In the sample survey for a highly skewed population, stratum jumpers often occur. Stratum jumpers are units having large discrepancies between a stratification variable and a study variable. We propose two models for stratum jumpers: a multiplicative model and a random replacement model. We also consider the modification of the L-H stratification algorithm such that we apply the previous models to L-H algorithm in determination of the sample sizes and the stratum boundaries. We evaluate the performances of the new stratification algorithms using real data. The result shows that L-H algorithm for the random replacement model outperforms other algorithms since the estimator has the least coefficient of variation.

• 한국의류학회지
• /
• 제29권6호
• /
• pp.885-895
• /
• 2005

The purpose of this study was to find out the most pursuasive social stratification variables affecting the attitude toward foreign luxury goods and the purchase frequency and examined the difference in the attitude toward foreign luxury goods and the purchase frequency among groups divided by age and social stratification variables. The subjects were 521 married women over 25 years old living in Seoul and Kyong-gi province areas. The data were analyzed by multiple regression, ANOVA, Duncan's multiple range test, frequency and percentage. Social stratification was measured by family's monthly income, educational and occupational levels of married women's and their husbands' for an objective method while economic levels, social status, consumption levels, and cultural levels were used fer a subjective method. The results were as follows; first, the most explainable variables influencing the attitude toward foreign luxury goods and the purchase frequency were age and women's educational levels examined by the objective method of social stratification. Second, according to the subjective method of social stratification, the attitude toward foreign luxury goods and the purchase frequency were affected by age, economic levels, consumption levels, and cultural levels. Consumption levels which showed actual expense per family were the most explainable variable in the purchase frequency.

• Journal of applied mathematics & informatics
• /
• 제26권1_2호
• /
• pp.161-176
• /
• 2008

The effect of variable viscosity on MHD mixed convection Hiemenz flow over a thermally stratified porous wedge plate has been studied in the presence of suction or injection. The wall of the wedge is embedded in a uniform Darcian porous medium in order to allow for possible fluid wall suction or injection and has a power-law variation of the wall temperature. An approximate numerical solution for the steady laminar boundary-layer flow over a wall of the wedge in the presence of thermal diffusion has been obtained by solving the governing equations using numerical technique. The fluid is assumed to be viscous and incompressible. Numerical calculations are carried out for different values of dimensionless parameters and an analysis of the results obtained shows that the flow field is influenced appreciably by the magnetic effect, variable viscosity, thermal stratification and suction / injection at wall surface. Effects of these major parameters on the transport behaviors are investigated methodically and typical results are illustrated to reveal the tendency of the solutions. Comparisons with previously published works are performed and excellent agreement between the results is obtained.

• 한국임상보건과학회지
• /
• 제1권3호
• /
• pp.49-61
• /
• 2013

Purpose. The purpose of this study is to analyze difference in quality of life related to health according to our country's social stratification. Methods. This study had analytical subjects as 7,992 adults(4,557 men, 3,435 women), who are included in a variable by social stratification among 15,691 people in the age group of 20~69 years old as examinees, based on the 4th Korea National Health and Nutrition Examination Survey. Results. This study obtained the following results. First, seeing the score of health-related life quality by social stratification, namely, the distribution of EQ-5D index, the new middle class(class II) was the highest(men with 0.966 point, women with 0.955 point). The agricultural self-employed class(class V) was indicated to be the lowest(men with 0.918 point, women with 0.866 point). In general, the more belonging to low social stratification led to having shown the lower aspect in EQ-5D index value. Second, as for the results of the hierarchical regression analysis, the factors of contributing to difference by social stratification in health-related quality of life were chronic disease, job stress, education, and income level in both men and women. The health behavior was grasped to be a factor of contributing only to women. Conclusions. To reduce difference in health-related quality of life according to social stratification in the future, the continuous investment is needed for supporting socially high-quality education and economic stability. The sphere of health will need to be given priority to the prevention of chronic disease and the development in effective management policy for the lower classes.

• 한국조사연구학회지:조사연구
• /
• 제11권3호
• /
• pp.103-114
• /
• 2010

계속조사를 위한 표본설계를 중간에 변경하려고 할 때 부딪히게 되는 문제 중 하나는 기존의 조사에서 중요하게 고려해야 할 주요 관심변수들에 대한 모집단 정보를 얻기 어렵다는 사실이다. 본 연구의 목적은 국민여행실태조사를 위한 표본설계에서 기존의 표본조사 자료와 새로운 추출틀 자료를 통합하여 효과적인 층화변수를 선정하는 방법을 제시하는 데 있다. 먼저 기존의 표본자료의 단위와 추출틀 단위 사이의 식별번호가 일치하지 않을 경우 번호 외에 일치시킬 수 있는 방안을 찾고, 이어 회귀분석과 같은 통계적 방법에 의해 변수 간의 연관성을 검토함으로써 적절한 층화변수를 선정하는 방법을 제시한다.

• 대한산업공학회지
• /
• 제36권1호
• /
• pp.32-41
• /
• 2010

DEA(Data Envelopment Analysis) is the relative efficiency measure among homogeneous DMU(Decision- Making Units) which can be used to useful tool to improve performance through efficiency evaluation and benchmarking. However, the general case of DEA was considered as unrealistic since it consists a benchmarking regardless of DMU characteristic by input and output elements and the high efficiency gap in benchmarking for inefficient DMU. To solve this problem, stratification method for benchmarking was suggested, but simply presented benchmarking path in repeatedly applying level. In this paper, we suggest a new method that inefficient DMU can choice the optimal path to benchmark the most efficient DMU base on the similarity among the input elements. For this, we propose a route choice method that combined a stratification benchmarking algorithm and SOM (Self-Organizing Map). An implementation on real environment is also presented.

• 한국의류학회지
• /
• 제23권3호
• /
• pp.459-470
• /
• 1999

The objectives of this study were to investigate the difference the clothing shopping orientation and clothing satisfaction according to satisfaction·dissatisfaction group to examine how the clothing satisfaction was influenced by consumer values demographic variable clothing shopping orientation. The subject were 457 women in Seoul Korea The results of the study were as follows. 1. five factors of clothing shopping orientation (SO) derived by factor analysis : F.1. conspicious SO : F,2 search SO: F,3 recreational SO : F,4 addictive SO :F,5 independent SO . Two factors of terminal value derived by factor analysis : F,1 responsible : F.2 ambitious. 2. Satisfaction group had high levels of search SO, dissatisfaction group had high levels of addictive SO. Satisfaction group was satisfied with color style appropriateness for wearer in order dissatisfaction group was dissatisfied with care price size in order. 3. Conspicious SO were influenced bysocial stratification social recognition and happiness. Search SO were influenced by dwelling area and age. Recreational SO were influenced by social stratification social recognition and responsible value. Addictive SO influenced by responsible value social recognition and happiness. independent SO were influenced by marital status and ambitious value. 4. Clothing satisfaction was influenced by addictive conspicious SO happiness and recreational SO(R2=24.6)

• Journal of the Korean Data and Information Science Society
• /
• 제28권2호
• /
• pp.443-449
• /
• 2017

Estimation of population proportion like the distribution rate of LED TV and the prevalence of a disease are often estimated based on survey sample data. Population proportion is generally considered as a special form of population mean. In complex sampling like stratified multistage sampling with unequal probability sampling, the denominator of mean may be random variable and it is estimated like ratio estimator. In this research, we examined the estimation of distribution rate based on stratified multistage sampling, and determined some numerical outcomes using stratified random sample data with about 25% of missing observations. In the data used for this research, the survey weight was determined by deterministic way. So, the weights are not random variable, and the population distribution rate and its variance estimator can be estimated like population mean estimation. When the weights are not random variable, if one estimates the variance of proportion estimator using ratio method, then the variances may be inflated. Therefore, in estimating variance for population proportion, we need to examine the structure of data and survey design before making any decision for estimation methods.

• 응용통계연구
• /
• 제14권2호
• /
• pp.289-304
• /
• 2001

2001년에 실시할 $\ulcorner$국민건강.영양조사$\lrcorner$를 위한 새로운 표본설계를 하였다. 본 표본설계에서는 표본의 대표성을 높이기 위해서 기존의 표본가구수는 유지하면서 표본조사구를 증가시키고 새로운 지역 층화변수를 추가로 도입하였다. 또한 추정량의 추정오차 공식을 유도하여 추정의 신뢰성을 측정할 수 있도록 설계하였다.

• 대한의생명과학회지
• /
• 제22권1호
• /
• pp.9-17
• /
• 2016

Unfortunately, the five-year survival rate of lung cancer is relatively low compared with other cancers. Therefore, better predictors are need for prognosis, therapeutic strategy, risk stratification and predicting long-term mortality of lung cancer. Recently, increasing data suggest that Glasgow Prognostic Score (GPS) and procalcitonin levels are useful predictor cancer prognosis. In this study, we retrospectively investigated the correlation of GPS or procalcitonin to clinical variables in patients with pretreatment lung cancer. In 135 patients with pretreatment lung cancer, GPS, procalcitonin, demographic characteristics, hematological, coagulation, biochemical, inflammatory and cardiac markers were measured. Monocyte, eosinophil, basophil, neutrophil to lymphocyte ratio, red cell distribution width (RDW), platelet to lymphocyte ratio, mean platelet volume to platecrit ratio, D-dimer and prothrombin time (PT) levels were higher, whereas mean platelet volume was lower than their normal ranges. Glucose and sodium levels were low, whereas gamma glutamyl transferase (GGT), total bilirubin, creatinine and inorganic phosphorus concentrations were increase compared their normal ranges. Procalcitonin, high sensitivity C-reactive protein and troponin-I concentrations were elevated compared with their normal ranges. GPS had significantly positive or negative relations to cancer stage, hematological, coagulation, biochemical, inflammatory and troponin-I. Based on the data, we suggest that GPS may be a potent and useful predictor for prognosis, therapeutic strategy, risk stratification and predicting long-term mortality of lung cancer.