Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Korean Journal of Applied Statistics
Journal Basic Information
Journal DOI :
The Korean Statistical Society
Editor in Chief :
Volume & Issues
Volume 18, Issue 3 - Nov 2005
Volume 18, Issue 2 - Jul 2005
Volume 18, Issue 1 - Mar 2005
Selecting the target year
Statistical Tests and Applications for the Stability of an Estimated Cointegrating Vector
Kim, Tae-Ho ; Hwang, Sung-Hye ; Kim, Mi-Yun ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 503~519
DOI : 10.5351/KJAS.2005.18.3.503
Cointegration test is usually performed under the assumption that the cointegrating vector is constant for the whole sample period. Most previous studies have used conventional cointegration methods in testing for a stable long-run equilibrium relation among related variables. However they have overlooked that the long-run equilibrium may not the unique and the stable relation may not be guaranteed. This study develops the additional statistical tests for the stability of the estimated cointegrating vector. Three tests for the parameter stability of a cointegrated regression model are utilized and applied to identify the types of variations in the long-run relation between the domestic unemployment and the rotated macroeconomic variables of interest. The present paper finds that, there exists a stable but, time-varying long-run relation between those. The observed variation in cointegrating relations is generally characterized by a discrete one-time shift, rather than a gradually evolving random walk process which is attributable to the IMF financial and economic crisis.
A Study on an Alternative to the Standardized Scoring System in CSAT
Hwang, Hyung-Tae ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 521~532
DOI : 10.5351/KJAS.2005.18.3.521
In the current College Scholastic Aptitude Test (CSAT), the standardized scoring system is being adopted to adjust te differences of degrees of difficulty between the optional subjects. But it becomes clear that the system has several weak points, some of which are considered to be very serious. In this paper we propose an alternative method, so-called the additive scoring system. It determines the additional points per each subjects, according to the subject mean scores. The proposed method has been simulated using the data of 2005' CSAT, and it turns out that the additive scoring system reduces or remove the troubles caused by the standardized scoring system.
Bayesian Model Selection for Linkage Analyses: Considering Collinear Predictors
Suh, Young-Ju ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 533~541
DOI : 10.5351/KJAS.2005.18.3.533
We identify the correct chromosome and locate the corresponding markers close to the QTL in the linkage analysis of a quantitative trait by using the SSVS method. We consider several markers linked to the QTL, as well as to each oyher and thus the i.b.d. values at these loci generate collinear predictors to be evaluated when using the SSVS approach. The results on considering only closely linked markers to two QTL simultaneously showed clear evidence in favor of the closest marker to the QTL considered over other markers. The results of the analysis of collinear markers with SSVS showeed high concordance to those obtained using traditional multiple regression. We conclude based on this simulation study that the SSVS is quite useful to identify linkage with multiple linked markers simultaneously for a complex quantitative trait.
Credit Scoring Using Splines
Koo Ja-Yong ; Choi Daewoo ; Choi Min-Sung ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 543~553
DOI : 10.5351/KJAS.2005.18.3.543
Linear logistic regression is one of the most widely used method for credit scoring in credit risk management. This paper deals with credit scoring using splines based on Logistic regression. Linear splines and an automatic basis selection algorithm are adopted. The final model is an example of the generalized additive model. A simulation using a real data set is used to illustrate the performance of the spline method.
A Development of Customer Segmentation by Using Data Mining Technique
Jin Seo-Hoon ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 555~565
DOI : 10.5351/KJAS.2005.18.3.555
To Know customers is very important for the company to survive in its cut-throat competition among coimpetitors. Companies need to manage the relationship with each ana every customer, ant make each of customers as profitable as possible. CRM (Customer relationship management) has emerged as a key solution for managing the profitable relationship. In order to achieve successful CRM customer segmentation is a essential component. Clustering as a data mining technique is very useful to build data-driven segmentation. This paper is concerned with building proper customer segmentation with introducing a credit card company case. Customer segmentation was built based only on transaction data which cattle from customer's activities. Two-step clustering approach which consists of k-means clustering and agglomerative clustering was applied for building a customer segmentation
An Application of gCRM Using Customer Information
Lee Sun-Soon ; Lee Hong-Seok ; Lee Joong-Hwan ; Kim Sung-Soo ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 567~581
DOI : 10.5351/KJAS.2005.18.3.567
Geographical Customer Relationship Management (gCRM) is an integrated solution of Geographic Information System (GIS) and Customer Relationship Management (CRM). In gCRM, GIS is used to show multi-dimensional analytical results of customer information geographically. When customer information is geographically presented, more valuable information appears. In this research we briefly introduce gCRM and show real examples of customer segmentation applied to company.
Applying Randomization Tests to Collocation Analyses in Large Corpora
Yang Kyung-Sook ; Kim HeeYoung ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 583~595
DOI : 10.5351/KJAS.2005.18.3.583
Contingency tables are used to compare counts of n-grams to determine if the n-gram is a true collocation, meaning that the words that make up the n-gram are highly associated in the text. Some statistical methods for identifying collocation are used. They are Kulczinsky coefficient, Ochiai coefficient, Frager and McGowan coefficient, Yule coefficient, mutual information, and chi-square, and so on. But the main problem is that these measures are based ell the assumption of a nor-mal or approximately normal distribution of the variables being sampled. While this assumption is valid in most instances, it is not valid when comparing the rates of occurrence of rare events, and texts are composed mostly of rare events. In this paper we have simply reviewed some statistics about testing association of two words. Some randomization tests to evaluate the significance level in analyzing collocation in large corpora are proposed. A related graph can be used to compare different lest statistics that ran be used to analyze the same contingency table.
Imputation for Binary or Ordered Categorical Traits Based on the Bayesian Threshold Model
Lee Seung-Chun ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 597~606
DOI : 10.5351/KJAS.2005.18.3.597
The nonresponse in sample survey causes a problem when it comes time to analyze dataset in public-use files where the user has only complete-data methods available and has limited information about the reasons for nonresponse. Recently imputation for nonresponse is becoming a standard approach for handling nonresponse and various imputation methods have been devised . However, most imputation methods concern with continuous traits while many interesting features are measured by binary or ordered categorical scales in sample survey. In this note. an imputation method for ignorable nonresponse in binary or ordered categorical traits is considered.
Interval Estimation for a Binomial Proportion Based on Weighted Polya Posterior
Lee Seung-Chun ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 607~615
DOI : 10.5351/KJAS.2005.18.3.607
Recently the interval estimation of a binomial proportion is revisited in various literatures. This is mainly due to the erratic behavior of the coverage probability of the will-known Wald confidence interval. Various alternatives have been proposed. Among them, Agresti-Coull confidence interval has been recommended by Brown et al. (2001) with other confidence intervals for large sample, say n
40. On the other hand, a noninformative Bayesian approach called Polya posterior often produces statistics with good frequentist's properties. In this note, an interval estimator is developed using weighted Polya posterior. The resulting interval estimator is essentially the Agresti-Coull confidence interval with some improved features. It is shown that the weighted Polys posterior produce an effective interval estimator for small sample size and a severely skewed binomial distribution.
Measures for Evaluating the Orthogonal Array of Strength 3
Jang Dae-Heung ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 617~625
DOI : 10.5351/KJAS.2005.18.3.617
We usually use orthogonal designs-orthogonal array of strength 2 as orthogonal arrays. It was shown that fractional factorial plaits represented by orthogonal arrays of strength 3 are universally optimal under the additive motels that includes the mean, all main effects and all two-factor interactions. Therefore, we need the measure for evaluating the orthogonal array of strength 3. We can extend this measure as the measure for evaluating the orthogonal array of strength t(
Space Time Autoregressive Model for Small Area Estimation
Kim Jae Doo ; Shin Key-Il ; Lee Sang Eun ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 627~637
DOI : 10.5351/KJAS.2005.18.3.627
Small area estimation has been studied using various methods such as direct, indirect, synthetic and based on regression or time series model . In this paper we investigate a motel-based small area estimation which takes into account the spare time autoregressive model. The Economic Active Population Surveys in 2001 are used for analysis and the results from space-time autoregressive(STAR) and simultaneous autoregressive(SAR) model are compared with using MSE, MAE and MB.
A Comparison of Two Models for Forecasting Mortality in South Korea
Park Yousung ; Kim Kee Whan ; Lee Dong-Hee ; Lee Yeon Kyung ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 639~654
DOI : 10.5351/KJAS.2005.18.3.639
The Lee and Carter method has widely used to forecast mortality because of the simple structure of model and the stable forecasting. The Lee and Carter method, however, also has limitations. The assumption of the rate of decline in mortality at each age remaining invariant over time has been violated in several decades. And, there is no way to include covariates in the model for better forecasts. Here we introduce Park, Choi and Kim method to make up for Lee and Carter's weak points by using two random processes. We discuss structural features of two methods. furthermore, for each method, we forecast life expectancy for 2005 to 2050 using South Korea data and compare the results.
Improvement in Performing a Test for Additional Information
Kim, Myung-Geun ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 655~660
DOI : 10.5351/KJAS.2005.18.3.655
An iterative method that greatly reduces a burden of computation in performing Rao's test for additional information in two-group discriminant analysis is suggested. A numerical example is provided for illustration.
Tree-structured Clustering for Continuous Data
Huh Myung-Hoe ; Yang Kyung-Sook ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 661~671
DOI : 10.5351/KJAS.2005.18.3.661
The aim of this study is to propose a clustering method, called tree-structured clustering, by recursively partitioning continuous multivariate dat a based on overall
criterion with a practical node-splitting decision rule. The clustering method produces easily interpretable clustering rules of tree types with the variable selection function. In numerical examples (Fisher's iris data and a Telecom case), we note several differences between tree-structured clustering and K-means clustering.
Design-based and model-based Inferences in Survey Sampling
Kim Kyu-Seong ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 673~687
DOI : 10.5351/KJAS.2005.18.3.673
We investigate both the design-based and model-based inferences, which are usual inferential methods in survey sampling. While the design-based inference is on the basis of randomization principle, The motel-based inference is based on likelihood principle as well as conditionality principle. There have been some disputes between two inferences for a long time and those have not yet been determined. In this paper we reviewed some issues on two inferences and compared their advantages and disadvantages in some viewpoints.
A Study on Sample Variance
Jang Dae-Heung ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 689~699
DOI : 10.5351/KJAS.2005.18.3.689
We usually use
as sample variance. Korean high school text-books use
as sample variance. We can compare the above two definitions of sample variance through their theoretical relationship and simulation.
Suppression for Logistic Regression Model
Hong C. S. ; Kim H. I. ; Ham J. H. ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 701~712
DOI : 10.5351/KJAS.2005.18.3.701
The suppression for logistic regression models has been debated no longer than that for linear regression models since, among many other reasons, sum of squares for regression (SSR) or coefficient of determination (
) could be defined into various ways. Based on four kinds of
's: two kinds are most preferred, and the other two are proposed by Liao & McGee (2003), four kinds of SSR's are derived so that the suppression for logistic models is explained. Many data fitted to logistic models are generated by Monte Carlo method. We explore when suppression happens, and compare with that for linear regression models.
Implementation of Questionnaire and Customer Satisfaction Investigation System on Internet
Namkung, Pyong ;
Korean Journal of Applied Statistics, volume 18, issue 3, 2005, Pages 713~727
DOI : 10.5351/KJAS.2005.18.3.713
The advantage of an internet survey is the speed with which data can be accumulated from respondents. And this method is more economical, provides more accurate information, and has greater scope in subject coverage. Since there is used multi-media, the design of questionnaires is even more important in order to achieve high data quality.