Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Communications for Statistical Applications and Methods
Journal Basic Information
Journal DOI :
The Korean Statistical Society
Editor in Chief :
Volume & Issues
Volume 10, Issue 3 - Dec 2003
Volume 10, Issue 2 - Aug 2003
Volume 10, Issue 1 - Apr 2003
Selecting the target year
Estimation of Geometric Mean for k Exponential Parameters Using a Probability Matching Prior
Kim, Hea-Jung ; Kim, Dae Hwang ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 1~9
DOI : 10.5351/CKSS.2003.10.1.001
In this article, we consider a Bayesian estimation method for the geometric mean of
exponential parameters, Using the Tibshirani's orthogonal parameterization, we suggest an invariant prior distribution of the
parameters. It is seen that the prior, probability matching prior, is better than the uniform prior in the sense of correct frequentist coverage probability of the posterior quantile. Then a weighted Monte Carlo method is developed to approximate the posterior distribution of the mean. The method is easily implemented and provides posterior mean and HPD(Highest Posterior Density) interval for the geometric mean. A simulation study is given to illustrates the efficiency of the method.
Power Comparison of EGLS Test Statistic for Fixed Effects with Arbitrary Distributions
Lee, Jang-Taek ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 11~18
DOI : 10.5351/CKSS.2003.10.1.011
Quite often normality assumptions are not satisfied in practical applications. In this paper, an estimated generalized least squares(EGLS) analysis are considered in two way mixed linear models with arbitrary types of distributions for random effects. We investigate the power performance of EGLS analysis based on Henderson's method III, ML, REML and MINQUE(1). The power performances depend on the imbalance of design, on the actual values of ratio of variance components, and on the skewness and kurtosis parameters of the underlying distributions slightly. Results of our limited simulation study suggest that the EGLS F-statistics using four estimators and arbitrary distributions produce similar type I error rates and power performance.
A Space-Time Model with Application to Annual Temperature Anomalies;
Lee, Eui-Kyoo ; Moon, Myung-Sang ; Gunst, Richard F. ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 19~30
DOI : 10.5351/CKSS.2003.10.1.019
Spatiotemporal statistical models are used for analyzing space-time data in many fields, such as environmental sciences, meteorology, geology, epidemiology, forestry, hydrology, fishery, and so on. It is well known that classical spatiotemporal process modeling requires the estimation of space-time variogram or covariance functions. In practice, the estimation of such variogram or covariance functions are computationally difficult and highly sensitive to data structures. We investigate a Bayesian hierarchical model which allows the specification of a more realistic series of conditional distributions instead of computationally difficult and less realistic joint covariance functions. The spatiotemporal model investigated in this study allows both spatial component and autoregressive temporal component. These two features overcome the inability of pure time series models to adequately predict changes in trends in individual sites.
Detection of Change-Points by Local Linear Regression Fit;
Kim, Jong Tae ; Choi, Hyemi ; Huh, Jib ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 31~38
DOI : 10.5351/CKSS.2003.10.1.031
A simple method is proposed to detect the number of change points and test the location and size of multiple change points with jump discontinuities in an otherwise smooth regression model. The proposed estimators are based on a local linear regression fit by the comparison of left and right one-side kernel smoother. Our proposed methodology is explained and applied to real data and simulated data.
Food and Agriculture Statistics of the UN-FAO
Kim, Joo-Hwan ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 39~48
DOI : 10.5351/CKSS.2003.10.1.039
Quality in agriculture statistics at the international level is very important to all countries in the world. We investigate international organizations which produce worldwide agricultural statistics. The UN-FAO is one of the most important organization in agricultural statistics fields. We first introduce the Statistics division in FAO and present some considerations like production, reliability, quality of the agriculture data from my own experience as a consultant in FAO and Ministry of Agriculture and Forestry in Korea.
A Study on Clustering Kansei Factors for the Surface Roughness of Materials
Jun, Chang Lim ; Choi, Kyungmee ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 49~60
DOI : 10.5351/CKSS.2003.10.1.049
The human sensibility product design requires information on consumer's emotions such as vision, auditory, olfactory, gustatory, or tactile perceptions. In this study, tactile sense which has not been well studied compared to other senses, is measured and statistically analysed. The emotional responses of 37 pairs of positive and negative adjectives describing tactile senses are collected and analysed through the questionnaire to find the correlation between adjectives and surface roughness of the sample. Mean ranks for 37 pairs of adjectives on four samples are obtained, and used to cluster these adjectives by factor analysis, multidimensional scaling, or cluster analysis.
Web Log Analysis Using Support Vector Regression
Jun, Sung-Hae ; Lim, Min-Taik ; Jorn, Hong-Seok ; Hwang, Jin-Soo ; Park, Seong-Yong ; Kim, Jee-Yun ; Oh, Kyung-Whan ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 61~77
DOI : 10.5351/CKSS.2003.10.1.061
Due to the wide expansion of the internet, people can freely get information what they want with lesser efforts. However without adequate forms or rules to follow, it is getting more and more difficult to get necessary information. Because of seemingly chaotic status of the current web environment, it is sometimes called "Dizzy web" The user should wander from page to page to get necessary information. Therefore we need to construct system which properly recommends appropriate information for general user. The representative research field for this system is called Recommendation System(RS), The collaborative recommendation system is one of the RS. It was known to perform better than the other systems. When we perform the web user modeling or other web-mining tasks, the continuous feedback data is very important and frequently used. In this paper, we propose a collaborative recommendation system which can deal with the continuous feedback data and tried to construct the web page prediction system. We use a sojourn time of a user as continuous feedback data and combine the traditional model-based algorithm framework with the Support Vector Regression technique. In our experiments, we show the accuracy of our system and the computing time of page prediction compared with Pearson's correlation algorithm.algorithm.
A Comparison of Variance Lower Bound between the Optimum Allocation and the Power Allocation
Son, Chang-Kyoon ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 79~88
DOI : 10.5351/CKSS.2003.10.1.079
In this paper, we study the efficiency of the stratified estimator in related with the variance lower bound of Horvitz-Thompson estimator subject to the superpopulation model. Especially, we compare the variance lower bound of optimum allocation with that of power allocation subject to Dalenius-Hedges stratification.
An Empirical Study on the Wealth Effect
Kim, Yon Hyong ; Chong, Young Suk ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 89~99
DOI : 10.5351/CKSS.2003.10.1.089
The primary purpose of this paper is to estimate the wealth effect. We establish a linear relationships between household consumption, labor income, and stock price index. Each variable is nonstationary. And so, it contains a unit root. However, as the result of the test about cointegrating relations, the variables are cointegrated which implies the error term is stationary. The cointegrating parameter that the marginal propensity to consume out of stock price is 0.08%. The result of estimation shows the error correction is -0.62 and the significant level is also high. The error correction term indicates a rather rapid adjustment to deviations from the long run equilibrium relations.
Efficient Quasi-likelihood Estimation for Nonlinear Time Series Models and Its Application
Kim, Sahmyeong ; Cha, Kyungyup ; Lee, Sungduck ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 101~113
DOI : 10.5351/CKSS.2003.10.1.101
Quasi likelihood estimators defined by Wedderburn are derived for several nonlinear time series models. And also, the least squared estimator and Quasi-likelihood estimator are compared in sense of asymptotic relative efficiency at those models. Finally, we apply these estimations to a real data on exchanging rate and stock market prices.
Usefulness of AB/BA/AA/BB Crossover Design
Nam, Jusun ; Kim, Dongjae ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 115~123
DOI : 10.5351/CKSS.2003.10.1.115
In this paper, we discuss the usefulness of AB/BA/AA/BB crossover design sense of sample size. The result is that the AB/BA/AA/BB crossover design is more economical than AB/BA one if large carryover effect exists and so is it in the comparison with completely randomized design within-subject correlation is large.
A Comparison of Influence Diagnostics in Linear Mixed Models
Lee, Jang-Taek ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 125~134
DOI : 10.5351/CKSS.2003.10.1.125
Standard estimation methods for linear mixed models are sensitive to influential observations. However, tools and concepts for linear mixed model diagnostics are rudimentary until now and research is heavily demanded in linear mixed models. In this paper, we consider two diagnostics to evaluate the effects of individual observations in the estimation of fixed effects for linear mixed models. Those are Cook's distance and COVRATIO. Results of our limited simulation study suggest that the Cook's distance is not good statistical quantity in linear mixed models. Also calibration point for COVRATIO seems to be quite conservative.
Empirical Comparisons of Disparity Measures for Partial Association Models in Three Dimensional Contingency Tables
Jeong, D.B. ; Hong, C.S. ; Yoon, S.H. ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 135~144
DOI : 10.5351/CKSS.2003.10.1.135
This work is concerned with comparison of the recently developed disparity measures for the partial association model in three dimensional categorical data. Data are generated by using simulation on each term in the log-linear model equation based on the partial association model, which is a proposed method in this paper. This alternative Monte Carlo methods are explored to study the behavior of disparity measures such as the power divergence statistic I(λ), the Pearson chi-square statistic X
, the likelihood ratio statistic G
, the blended weight chi-square statistic BWCS(λ), the blended weight Hellinger distance statistic BWHD(λ), and the negative exponential disparity statistic NED(λ) for moderate sample sizes. We find that the power divergence statistic I(2/3) and the blended weight Hellinger distance family BWHD(1/9) are the best tests with respect to size and power.
Unified Estimations for Parameter Changes in the Uniform Distribution
Lee, Changsoo ; Chang, Chuseock ; Park, Yangwoo ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 145~151
DOI : 10.5351/CKSS.2003.10.1.145
We shall propose several estimators for the scale parameter in the uniform distribution when the parameter is functions of a known exposure level, and obtain expectations and variances for their proposed estimators. And we shall compare numerically relative efficiencies for proposed estimators of the scale parameter in the small sample sizes.
Using Evolutionary Optimization to Support Artificial Neural Networks for Time-Divided Forecasting: Application to Korea Stock Price Index
Oh, Kyong Joo ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 153~166
DOI : 10.5351/CKSS.2003.10.1.153
This study presents the time-divided forecasting model to integrate evolutionary optimization algorithm and change point detection based on artificial neural networks (ANN) for the prediction of (Korea) stock price index. The genetic algorithm(GA) is introduced as an evolutionary optimization method in this study. The basic concept of the proposed model is to obtain intervals divided by change points, to identify them as optimal or near-optimal change point groups, and to use them in the forecasting of the stock price index. The proposed model consists of three phases. The first phase detects successive change points. The second phase detects the change-point groups with the GA. Finally, the third phase forecasts the output with ANN using the GA. This study examines the predictability of the proposed model for the prediction of stock price index.
Simplicial Regression Depth with Censored and Truncated Data
Park, Jinho ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 167~175
DOI : 10.5351/CKSS.2003.10.1.167
In this paper we develop a robust procedure to estimate regression coefficients for a linear model with censored and truncated data based on simplicial regression depth. Simplicial depth of a point is defined as the proportion of data simplices containing it. This simplicial depth can be extended to regression problem with censored and truncated data. Any line can be given a depth and the deepest regression line is the line with the maximum simplicial regression depth. We show how the proposed regression performs through analyzing AIDS incubation data.
Characterization of Some Classes of Distributions Related to Operator Semi-stable Distributions
Joo, Sang Yeol ; Yoo, Young Ho ; Choi, Gyeong Suk ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 177~189
DOI : 10.5351/CKSS.2003.10.1.177
For a positive integer m, operator m-semi-stability and the strict operator m-semi-stability of probability measures on
are defined. The operator m-semi-stability is a generalization of the definition of operator semi-stability with exponent Q. Characterization of strictly operator na-semi-stable distributions among operator m-semi-stable distributions is given. Translation of strictly operator m-semi-stable distribution is discussed.
Some Results on Availability of Repairable Component and Repairable Coherent System
Cha, Ji-Hwan ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 191~202
DOI : 10.5351/CKSS.2003.10.1.191
Availability is an important measure of performance of a repairable component. In this paper, the explicit expression for the availability of a repairable component, which is subject to the policy II(Age Replacement Policy) of Barlow and Hunter (1960), is obtained and the existence of the steady state availability is shown. The steady state availabilities of the model are also obtained for the cases when the mean of the minimal repair time is increasing at a geometric rate or linearly increasing, In order to show the importance and the utility of the obtained result, we also consider an illustrative example of the repairable coherent system whose components are repairable, and the obtained results are applied to derive the steady state availability of the whole system. In this situation, we can see that the condition of the existence of the steady state availability for each component is essential. Some remarks on the optimal replacement policy that maximizes the steady state availability are also given.
Bayesian Estimation of State-Space Model Using the Hybrid Monte Carlo within Gibbs Sampler
Park, Ilsu ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 203~210
DOI : 10.5351/CKSS.2003.10.1.203
In a standard Metropolis-type Monte Carlo simulation, the proposal distribution cannot be easily adapted to "local dynamics" of the target distribution. To overcome some of these difficulties, Duane et al. (1987) introduced the method of hybrid Monte Carlo(HMC) which combines the basic idea of molecular dynamics and the Metropolis acceptance-rejection rule to produce Monte Carlo samples from a given target distribution. In this paper, using the HMC within Gibbs sampler, an asymptotical estimate of the smoothing mean and a general solution to state space modeling in Bayesian framework is obtaineds obtained
Tests to Detect Changes in Micro-Flora Composition;
Kim, Donguk ; Yang, Mark C.K. ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 211~224
DOI : 10.5351/CKSS.2003.10.1.211
Good's lambda test, a permutation test used to detect the changes of microorganism composition under two pathological conditions, has been quite popular for studying the micro-flora responsible for periodontal disease. A vast number of different micro-flora in the mouth renders the traditional chi-square test inapplicable. The main purpose of this paper is to evaluate the power of this test so that the sample size can be determined at the design stage. The robustness of this test and its comparison to two other intuitive tests are also presented. It is found that a permutation test based on likelihood ratio is more powerful than the lambda test in our simulated cases.
Koo, Ja-Yong ; Park, Heon Jin ; Choi, Daewoo ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 225~232
DOI : 10.5351/CKSS.2003.10.1.225
The support vector machine (SVM) is becoming increasingly popular in classification. The import vector machine (IVM) has been introduced for its advantages over SMV. This paper tries to improve the IVM. The proposed method, which is referred to as the polychotomous machine (PM), uses the Newton-Raphson method to find estimates of coefficients, and the Rao and Wald tests, respectively, for addition and deletion of import points. Because the PM basically follows the same addition step and adopts the deletion step, it uses, typically, less import vectors than the IVM without loosing accuracy. Simulated and real data sets are used to illustrate the performance of the proposed method.
A Note on the Asymptotic Property of S
in Linear Regression Model with Correlated Errors
Lee, Seung-Chun ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 233~237
DOI : 10.5351/CKSS.2003.10.1.233
An asymptotic property of the ordinary least squares estimator of the disturbance variance is considered in the regression model with correlated errors. It is shown that the convergence in probability of S
is equivalent to the asymptotic unbiasedness. Beyond the assumption on the design matrix or the variance-covariance matrix of disturbances error, the result is quite general and simplify the earlier results.
Input Variable Importance in Supervised Learning Models
Huh, Myung-Hoe ; Lee, Yong Goo ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 239~246
DOI : 10.5351/CKSS.2003.10.1.239
Statisticians, or data miners, are often requested to assess the importances of input variables in the given supervised learning model. For the purpose, one may rely on separate ad hoc measures depending on modeling types, such as linear regressions, the neural networks or trees. Consequently, the conceptual consistency in input variable importance measures is lacking, so that the measures cannot be directly used in comparing different types of models, which is often done in data mining processes, In this short communication, we propose a unified approach to the importance measurement of input variables. Our method uses sensitivity analysis which begins by perturbing the values of input variables and monitors the output change. Research scope is limited to the models for continuous output, although it is not difficult to extend the method to supervised learning models for categorical outcomes.
The Three-Stage Cluster Randomized Response Model for Obtaining Sensitive Information
Lee, Gi Sung ; Hong, Ki Hak ; Son, Chang Kyoon ; Jung, Young Mee ;
Communications for Statistical Applications and Methods, volume 10, issue 1, 2003, Pages 247~256
DOI : 10.5351/CKSS.2003.10.1.247
In this study, we systemize the theoretical validity for applying RRM to three-stage cluster sampling method and derive the estimate and it's variance of sensitive parameter. We derive the minimum variance form under the optimal values of the subsample sizes when the costs are fixed. Under the some given precision, we obtain the optimal values of the subsample sizes and derive the minimum cost form by using them. We apply the three-stage cluster RRM to field survey and suggest some necessary points for practical use.