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 24, Issue 6 - Dec 2011
Volume 24, Issue 5 - Oct 2011
Volume 24, Issue 4 - Aug 2011
Volume 24, Issue 3 - Jun 2011
Volume 24, Issue 2 - Apr 2011
Volume 24, Issue 1 - Feb 2011
Selecting the target year
A Comparison of Statistical Prediction Models in Household Water End-Uses
Myoung, Sung-Min ; Lee, Doo-Jin ; Kim, Hwa-Soo ; Jo, Jin-Nam ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 567~573
DOI : 10.5351/KJAS.2011.24.4.567
This study develops a predictive model for household water end-uses based on data that have measured household characteristics, housing characteristics and other items, surveyed over 3 years in Korea. However, the measured data was left-skewed and it was not fitted to normal distribution. The parameter estimate were biased when using a multiple regression model. In addition, the results of the testing for the model were usually of significance due to the tiny residual from a large number of observations. In order to solve the problem, we suggested log-normal regression model and Weibull regression model as alternatives. The results of this study can be utilized in the planning stages of water and waste water facilities.
Ruin Probability on Insurance Risk Models
Park, Hyun-Suk ; Choi, Jeong-Kyu ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 575~586
DOI : 10.5351/KJAS.2011.24.4.575
In this paper, we study an asymptotic behavior of the finite-time ruin probability of the compound Poisson model in the case that the initial surplus is large. To compare an exact ruin probability with an approximate one, we place the focus on the exact calculation for the ruin probability when the claim size distribution is regularly varying tailed (i.e. exponential claims and inverse Gaussian claims). We estimate an adjustment coefficient in these examples and show the relationship between the adjustment coefficient and the safety premium. The illustration study shows that as the safety premium increases so does the adjustment coefficient. Larger safety premium means lower "long-term risk", which only stands to reason since higher safety premium means a faster rate of safety premium income to offset claims.
Generalized Partially Linear Additive Models for Credit Scoring
Shim, Ju-Hyun ; Lee, Young-K. ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 587~595
DOI : 10.5351/KJAS.2011.24.4.587
Credit scoring is an objective and automatic system to assess the credit risk of each customer. The logistic regression model is one of the popular methods of credit scoring to predict the default probability; however, it may not detect possible nonlinear features of predictors despite the advantages of interpretability and low computation cost. In this paper, we propose to use a generalized partially linear model as an alternative to logistic regression. We also introduce modern ensemble technologies such as bagging, boosting and random forests. We compare these methods via a simulation study and illustrate them through a German credit dataset.
Comparison of Dimension Reduction Methods for Time Series Factor Analysis: A Case Study
Lee, Dae-Su ; Song, Seong-Joo ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 597~607
DOI : 10.5351/KJAS.2011.24.4.597
Value at Risk(VaR) is being widely used as a simple tool for measuring financial risk. Although VaR has a few weak points, it is used as a basic risk measure due to its simplicity and easiness of understanding. However, it becomes very difficult to estimate the volatility of the portfolio (essential to compute its VaR) when the number of assets in the portfolio is large. In this case, we can consider the application of a dimension reduction technique; however, the ordinary factor analysis cannot be applied directly to financial data due to autocorrelation. In this paper, we suggest a dimension reduction method that uses the time-series factor analysis and DCC(Dynamic Conditional Correlation) GARCH model. We also compare the method using time-series factor analysis with the existing method using ordinary factor analysis by backtesting the VaR of real data from the Korean stock market.
A Finite Mixture Model for Gene Expression and Methylation Pro les in a Bayesian Framewor
Jeong, Jae-Sik ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 609~622
DOI : 10.5351/KJAS.2011.24.4.609
The pattern of methylation draws significant attention from cancer researchers because it is believed that DNA methylation and gene expression have a causal relationship. As the interest in the role of methylation patterns in cancer studies (especially drug resistant cancers) increases, many studies have been done investigating the association between gene expression and methylation. However, a model-based approach is still in urgent need. We developed a finite mixture model in the Bayesian framework to find a possible relationship between gene expression and methylation. For inference, we employ Expectation-Maximization(EM) algorithm to deal with latent (unobserved) variable, producing estimates of parameters in the model. Then we validated our model through simulation study and then applied the method to real data: wild type and hydroxytamoxifen(OHT) resistant MCF7 breast cancer cell lines.
Clustering of Time-Course Microarray Data Using Pharmacokinetic Parameter
Lee, Hyo-Jung ; Kim, Peol-A ; Park, Mi-Ra ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 623~631
DOI : 10.5351/KJAS.2011.24.4.623
A major goal of time-course microarray data analysis is the detection of groups of genes that manifest similar expression patterns over time. The corresponding numerous cluster algorithms for clustering time-course microarray data have been developed. In this study, we proposed a clustering method based on the primary pharmacokinetic parameters in the pharmacokinetics study for assessment of pharmaceutical equivalents between two drug products. A real data and a simulation data was used to demonstrate the usefulness of the proposed method.
Statistical Consideration of Vaccine Clinical Trials
Nam, Ju-Sun ; Kang, Seung-Ho ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 633~646
DOI : 10.5351/KJAS.2011.24.4.633
Clinical vaccines studies (that include cancer prevention vaccines and therapeutic vaccines) are ongoing to improve the quality of life and lengthen the human lifespan. Recently clinical trials and research on vaccines have become more active due to the prevalence of new viruses such as the A(H1N1) virus that freighted the whole word in 2009. In this paper we will describe the statistical aspects of clinical vaccine trials and outline the current situation of domestic and international vaccine development.
λ Matrix for Evaluating an Incomplete Bloc Design
Jang, Dae-Heung ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 647~656
DOI : 10.5351/KJAS.2011.24.4.647
Incidence matrix is a useful tool for presenting incomplete block designs; however, it is inadequate to use only an incidence matrix in examining whether a certain incomplete block design becomes a balanced incomplete block design or not. We can use a structural matrix as a useful tool to show whether a certain incomplete block design becomes a balanced incomplete block design or not. We propose an augmented incidence matrix and
matrix as another tools for evaluating incomplete block designs. Through the augmented incidenc matrix and
matrix, we can ascertain whether a certain incomplete block design becomes a balance incomplete block design or not.
Implementation of Markov Chain: Review and New Application
Park, Chang-Soon ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 657~676
DOI : 10.5351/KJAS.2011.24.4.657
Properties of statistical process control procedures may not be derived analytically in many cases; however, the application of a Markov chain can solve such problems. This article shows how to derive the properties of the process control procedures using the generated Markov chains when the control statistic satisfies the Markov property. Markov chain approaches that appear in the literature (such as the statistical design and economic design of the control chart as well as the variable sampling rate design) are reviewed along with the introduction of research results for application to a new control procedure and reset chart. The joint application of a Markov chain approach and analytical solutions (when available) can guarantee the correct derivation of the properties. A Markov chain approach is recommended over simulation studies due to its precise derivation of properties and short calculation times.
Bayesian Approaches to Zero Inflated Poisson Model
Lee, Ji-Ho ; Choi, Tae-Ryon ; Wo, Yoon-Sung ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 677~693
DOI : 10.5351/KJAS.2011.24.4.677
In this paper, we consider Bayesian approaches to zero inflated Poisson model, one of the popular models to analyze zero inflated count data. To generate posterior samples, we deal with a Markov Chain Monte Carlo method using a Gibbs sampler and an exact sampling method using an Inverse Bayes Formula(IBF). Posterior sampling algorithms using two methods are compared, and a convergence checking for a Gibbs sampler is discussed, in particular using posterior samples from IBF sampling. Based on these sampling methods, a real data analysis is performed for Trajan data (Marin et al., 1993) and our results are compared with existing Trajan data analysis. We also discuss model selection issues for Trajan data between the Poisson model and zero inflated Poisson model using various criteria. In addition, we complement the previous work by Rodrigues (2003) via further data analysis using a hierarchical Bayesian model.
Image Fusion Based on Statistical Hypothesis Test Using Wavelet Transform
Park, Min-Joon ; Kwon, Min-Jun ; Kim, Gi-Hun ; Shim, Han-Seul ; Lim, Dong-Hoon ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 695~708
DOI : 10.5351/KJAS.2011.24.4.695
Image fusion is the process of combining multiple images of the same scene into a single fused image with application to many fields, such as remote sensing, computer vision, robotics, medical imaging and military affairs. The widely used image fusion rules that use wavelet transform have been based on a simple comparison with the activity measures of local windows such as mean and standard deviation. In this case, information features from the original images are excluded in the fusion image and distorted fusion images are obtained for noisy images. In this paper, we propose the use of a nonparametric squared ranks test on the quality of variance for two samples in order to overcome the influence of the noise and guarantee the homogeneity of the fused image. We evaluate the method both quantitatively and qualitatively for image fusion as well as compare it to some existing fusion methods. Experimental results indicate that the proposed method is effective and provides satisfactory fusion results.
Multiple-Group Latent Transition Model for the Analysis of Sequential Patterns of Early-Onset Drinking Behaviors among U.S. Adolescents
Chung, Hwan ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 709~719
DOI : 10.5351/KJAS.2011.24.4.709
We investigate the latent stage-sequential patterns of drinking behaviors of U.S. adolescents who have started to drink by age 14 years (seven years before the legal drinking age). A multiple-group latent transition analysis(LTA) with logistic regression is employed to identify the subsequent patterns of drinking behaviors among early-onset drinkers. A sample of 1407 early-onset adolescents from the National Longitudinal Survey of Youth(NLSY97) is analyzed using maximum-likelihood estimation. The analysis demonstrates that early-onset adolescents' drinking behaviors can be represented by four latent classes and their prevalence and transition are influenced by demographic factors of gender, age, and race.
Direct Nonparametric Estimation of State Price Density with Regularized Mixture
Jeon, Yong-Ho ;
Korean Journal of Applied Statistics, volume 24, issue 4, 2011, Pages 721~733
DOI : 10.5351/KJAS.2011.24.4.721
We consider the state price densities that are implicit in financial asset prices. In the pricing of an option, the state price density is proportional to the second derivative of the option pricing function and this relationship together with no arbitrage principle imposes restrictions on the pricing function such as monotonicity and convexity. Since the state price density is a proper density function and most of the shape constraints are caused by this, we propose to estimate the state price density directly by specifying candidate densities in a flexible nonparametric way and applying methods of regularization under extra constraints. The problem is easy to solve and the resulting state price density estimates satisfy all the restrictions required by economic theory.