• 대한인간공학회지
• /
• 제24권4호
• /
• pp.39-46
• /
• 2005

The purpose of the study is to suggest a methodology for properly categorizing the body type of Koreans based on the multivariate statistical analysis. Anthropometric data used in the study were measured from the sampled strata of about fifteen thousand Koreans surveyed through the 5th national anthropometic data measurement project called Size Korea funded by ATS, Korea, during 2003-2004. In order to categorize whole body types, the normalized anthropometric variables, being divided by its stature, were used for obtaining a set of factors that supposedly represent body types through the factor analysis. These factors, which were again clustered, yielded the body types according to the gender. The body types classified are expected to be applied to product design for clothing, furniture, automobile packaging, etc.

• 전자공학회논문지D
• /
• 제34D권10호
• /
• pp.7-13
• /
• 1997

A modern electronic system located at a certain distance form the discharge may respond with unexpected sensitivity ot that phenomenon, even if the phenomenon is so slight as to have been ignored in the port. It has been found that electromagnetic wave energy is emitted as a results of this electrostatic discharge between metallic objects. In order to theoretically examine the peculiar phenomenon, we propose an analytical approach to model the indirect ESD effect. A soruce model is given here using the spark resistence presented by rompe-weizel. A model experiment for indirect eSD is also conducted to express ESD detection rate by the statistical estimation. We verify that the statistical estimations agree the theoretical curve resulted from the rompe-weisel resistence.

• 전기의세계
• /
• 제22권4호
• /
• pp.17-24
• /
• 1973

In the view point of practical engineering the identification problem may be considered as a problem to determine the optimal model in the sense of minimizing a given criterion function using the input-output records of the plant. In the system identification the statistical approach has been known to be very effective when the topological structure of the system and the statistical characteristics of the observation noises are known a priori. But in the practical situation there are many cases when the inforhation about the observation noises or the system noises are not available a priori. Here, the authors propose a new identification method which can be used effectively even in the cases when the variances of observation noises are unknown a priori. In the method, the identification of unknown parameters of a linear diserete system is achieved by minimizing the improved quadratic criterion function which is composed of the term of square equation errors and the term to eliminate the affection of observation noises. The method also gives the estimate of noise variance. Numerical computations for several examples show that the proposed procedure gives satisfactory results even when the short time observation data are provided.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 1996년도 추계학술대회발표논문집; 고려대학교, 서울; 26 Oct. 1996
• /
• pp.19-22
• /
• 1996

A traditional approach to the prediction of economic and financial variables takes the form of statistical models to summarize past observations and to project them into the envisioned future. Over the past decade, an increasing number of organizations has turned to the use of neural networks. To date, however, many spheres of interest still lack a systematic evaluation of the statistical and neural approaches. One of these lies in the prediction of corporate bond yields for Korea. This paper reports on a comparative evaluation of ARIMA models and neural networks in the context of interest rate prediction. An additional experiment relates to an integration of the two methods. More specifically, the statistical model serves as a filter by providing estimtes which are then used as input into the neural network models.

• 산업경영시스템학회지
• /
• 제27권2호
• /
• pp.114-122
• /
• 2004

In statistical process control applications, variable sample size (VSS) $\bar{X}$ chart is often used to detect the assignable cause quickly. However, it is usually assumed that only one assignable cause results in the out-of-control in the process. In this paper, we propose the algorithm to minimize the function of cost per unit time and compare the economic design and the statistical design by use of the value of cost per unit time. We consider double assignable causes to occur with compound in the process and adopt the Markov chain approach to investigate the statistical properties of VSS $\bar{X}$ chart. A procedure that can calculate the control chart's parameters is proposed by the economic design.

• Communications for Statistical Applications and Methods
• /
• 제16권2호
• /
• pp.373-381
• /
• 2009

Evidence of linkage is expressed as a decreasing trend of the squared trait difference of two siblings with increasing identical by descent scores. In contrast to successes in the application of a parametric approach of Haseman-Elston regression, notably low powers are demonstrated in the nonparametric linkage analysis methods for complex traits and diseases with sib-pairs data. We report that the Genehunter nonparametric linkage statistic is biased and furthermore the variance formula that they used is an inflated one, and this is one reason for a low performance. Thus, we propose bias-corrected nonparametric linkage statistics. Simulation studies comparing our proposed nonparametric test statistics versus the existing test statistics suggest that the bias-corrected new nonparametric test statistics are more powerful and attains efficiencies close to that of Haseman-Elston regression.

• Communications for Statistical Applications and Methods
• /
• 제27권3호
• /
• pp.269-284
• /
• 2020

This study compares two teaching methods in an introductory statistics course at a large state university. The first method is the traditional lecture-based approach. The second method implements a flipped classroom that incorporates the recommendations of the American Statistical Association's Guidelines for Assessment and Instruction in Statistics Education (GAISE) College Report. We compare these two methods, based on student performance, illustrate the procedures of the flipped pedagogy, and discuss the impact of aligning our course to current guidelines for teaching statistics at the college level. Results show that students in the flipped class performed better than students in traditional delivery. Student questionnaire responses also indicate that students in flipped delivery aligned with the GAISE recommendations have built a productive mindset in statistics.

• Journal of Biosystems Engineering
• /
• 제23권6호
• /
• pp.621-628
• /
• 1998

In order to develop an inspection algorithm for an automatic eggshell inspection system, acoustic impulse response with neural network method was studied. An improved error backpropagation algorithm was selected as a loaming rule of neural network, and three layer network was chosen for the neural network architecture. Acoustic signals in time domain and theirs power spectrum were studied as the input to the neural network. The classification feasibility and success rate were investigated in terms of statistical analysis and neural network approach. As a result, the success rate was 95% with the statistical model having five independent variables. Among the neural network models studied, the power spectrum of acoustic signal as the input with 64 input neurons and the two impact data showed the success rate of 95.5% which was slightly higher than of statistical analysis.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2005년도 추계 학술발표회 논문집
• /
• pp.55-58
• /
• 2005

This paper discusses the asymptotic efficiency of estimators for optimal portfolios when returns are vector-valued non-Gaussian stationary processes. We give the asymptotic distribution of portfolio estimators ${\hat{g}}$ for non-Gaussian dependent return processes. Next we address the problem of asymptotic efficiency for the class of estimators ${\hat{g}}$ First, it is shown that there are some cases when the asymptotic variance of ${\hat{g}}$ under non-Gaussianity can be smaller than that under Gaussianity. The result shows that non-Gaussianity of X(t) does not always affect worse. Second, we give a necessary and sufficient condition for ${\hat{g}}$ to be asymptotically efficient when the return process is Gaussian, which shows that ${\hat{g}}$ is not asymptotically efficient generally. From this point of view we propose to use maximum likelihood type estimators for g, which are asymptotically efficient. We examine our approach numerically.

• 한국통신학회논문지
• /
• 제35권2A호
• /
• pp.196-201
• /
• 2010

Measurement of Quality of Service (QoS) parameters and its statistical analysis becomes a key issue for Mobile WiMAX service providers to manage the converged network efficiently and to support end-to-end QoS. In this paper, we investigate the population distribution of end-to-end one-way delay which is the most important QoS parameter in Mobile WiMAX networks. The samples are analyzed with Chi-Square Goodness-of-Fit test, Kolmogorov-Smirnov (K-S), and Anderson-Darling (A-D) test to verify the distribution of parent population. The relation with confidence level and the minimum number of sample size is also performed for logistic distribution. The statistical analysis is a promising approach for measuring the performance Mobile WiMAX networks.