• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.361-368
• /
• 1993

In recent years, kernel type estimates are abundant. In this paper, we propose a bandwidth selection method for kernel regression of fixed design based on bootstrap procedure. Mathematical properties of proposed bootstrap-based bandwidth selection method are discussed. Performance of the proposed method for small sample case is compared with that of cross-validation method via a simulation study.

• Journal of the Korean Statistical Society
• /
• v.22 no.2
• /
• pp.325-337
• /
• 1993

This paper propose two test statistics which enable us to proceed the variable selection in Fisher's linear discriminant function for the case of heterogeneous discrimination with equal training sample size. Simultaneous confidence intervals associated with the test are also given. These are exact and approximate results. The latter is based upon an approximation of a linear sum of Wishart distributions with unequal scale matrices. Using simulated sampling experiments, powers of the two tests have been tabulated, and power comparisons have been made between them.

• Journal of the Korean Statistical Society
• /
• v.14 no.1
• /
• pp.39-47
• /
• 1985

In this paper, we propose some nonparametric subset selection procedures for scale parameters based on rank-likes. The proposed procedures are compared to the Gupta-Sobel's parametric prcedure through a small-sample Monte Carlo study. The results show that the nonparametric procedures are quite robust for heavy-tailed distributions, but they have somewhat low efficiencies.

• Journal of the Korean Statistical Society
• /
• v.16 no.1
• /
• pp.26-36
• /
• 1987

In this paper, we study on a subset selection procedure based on Hodges-Lehmann estimators derived from the Wilcoxon test. To estimate the standard error of the Hodges-Lehmann estimators, the biweight A-estimator of scale is used. The Pitman efficiency of the proposed rule is compared with the Gupta's rule and the trimmed-means rule through a small-sample Monte Carlo study. The results show that the proposed rule satisfies the $P^*$-condition and is very efficient in various heavy-tailed distributions.

• Journal of the Korean Statistical Society
• /
• v.16 no.1
• /
• pp.37-44
• /
• 1987

A two-stage selection procedure is considered in the case of exponential populations with common known scale parameter. The proposed procedure is designed following the lines of Tamhane and Bechhofer(1977). The design constants to implement the procedure are provided. Monte Carlo results show that the proposed procedure performs better than the single procedure by Raghvachari and Starr (1970) in terms of the expected total sample size.

• Journal of the Korean Society of Costume
• /
• v.50 no.4
• /
• pp.63-72
• /
• 2000

The purpose of this study was to investigate the effects of lifestyle factors on clothing purchase motives, information use, and selection criteria in male college students. The sample included 241 male college students, and an instrument was developed based on the previous studies. The statistical analyses used for this study were factor analysis and multiple regression. The result of factor analysis showed that lifestyle consisted of six factors : clothing interest, serif-confidence, social participation, planned clothing purchase, family-orientation, and conservativeness. Clothing purchase motives consisted of conspicuous consumption motives, fashion and individuality motives, and economic motives. Clothing information use consisted of four factors: paper/display, personal advice, fashion show/clothing observation, and electronic media. Finally. clothing selection criteria consisted of practicability, fashion/individuality, and conformity, Multip1e regression revealed that there were significant effects of lifestyle factors on clothing purchase motives, information use, and selection criteria. For example, self-confidence factor had a negative impact on conspicuous consumption motive, personal advice information use, and fashion/individuality criteria. The relative importance of lifestyle factors were different according to different dimensions of clothing purchase motives, information use, and selection criteria.

• ETRI Journal
• /
• v.27 no.6
• /
• pp.747-758
• /
• 2005

The problem of selection among competing models has been a fundamental issue in statistical data analysis. Good fits to data can be misleading since they can result from properties of the model that have nothing to do with it being a close approximation to the source distribution of interest (for example, overfitting). In this study we focus on the preference among models from a family of polynomial regressors. Three decades of research has spawned a number of plausible techniques for the selection of models, namely, Akaike's Finite Prediction Error (FPE) and Information Criterion (AIC), Schwartz's criterion (SCH), Generalized Cross Validation (GCV), Wallace's Minimum Message Length (MML), Minimum Description Length (MDL), and Vapnik's Structural Risk Minimization (SRM). The fundamental similarity between all these principles is their attempt to define an appropriate balance between the complexity of models and their ability to explain the data. This paper presents an empirical study of the above principles in the context of model selection, where the models under consideration are univariate polynomials. The paper includes a detailed empirical evaluation of the model selection methods on six target functions, with varying sample sizes and added Gaussian noise. The results from the study appear to provide strong evidence in support of the MML- and SRM- based methods over the other standard approaches (FPE, AIC, SCH and GCV).

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.34 no.s01
• /
• pp.26-32
• /
• 1989

The analytic principles of GC and MS were explained in relation to plant hormone analyses and the characteristics of two instruments were compared. The selection of column, condition of measurement and the method of ionization to get a good spectrum were also briefly described. Finally, the pre-treatment of sample by solvent extraction method to remove the unnecessary part of sample and the synthetic method, especially reagents and reaction condition, for the preparation of ether or ester derivative which can be easily vaporized in GC were explained.

• The Korean Journal of Applied Statistics
• /
• v.21 no.2
• /
• pp.301-312
• /
• 2008

Mutual information is a measure of association of explanatory variable for predicting target variable. It is used for variable ranking and variable subset selection. This study is about the Sample-spacing approach which can be used for the estimation of mutual information from data consisting of continuous explanation variables and categorical target variable without estimating a joint probability density function. The results of Monte-Carlo simulation and experiments with real-world data show that m = 1 is preferable in using Sample-spacing.

• Journal of Nutrition and Health
• /
• v.17 no.3
• /
• pp.236-241
• /
• 1984

In order to make clear the relationship between sample design and sample survey in community, it was conducted research on sample design for National Nutrition Survey in 1983. In this paper it was tried to analize the data based on The Report of a Settled Population, 1981 conducted by National Bureau of Statistics Economic Planning Board. The sample was basically using stratified two-stage sampling with systematic sampling of Ban or Li as administrative unit. The population represents the whole nation excluding Jeju-do because of budget. The selection of sampling unit and sampling procedure was as follows. 1) Stratify the nation-wide area in 20 sections according to administrative districts. 2) Determine the sample size in each section according to equal proportional rate (1 / 8040) and to about 1,000 households in the sample. 3) Select the 25 sampling units by section according to households proportion. 4) Select the 10 households at random from each Ban or Li according to equal probability proportion as the final sampling unit. Using the procedure, it was sampled 1,000 households for National Nutrition Survey in 1983.