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 28, Issue 6 - Dec 2015
Volume 28, Issue 5 - Oct 2015
Volume 28, Issue 4 - Aug 2015
Volume 28, Issue 3 - Jun 2015
Volume 28, Issue 2 - Apr 2015
Volume 28, Issue 1 - Feb 2015
Selecting the target year
Bayesian Analysis for Uncertainty of Radiocarbon Dating
Lee, Youngseon ; Lee, Jaeyong ; Kim, Jangsuk ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 371~383
DOI : 10.5351/KJAS.2015.28.3.371
Use of radiocarbon dating is increasing for chronology; however, its variability and discrepancy with existing chronologies can cause doubts in regards to credibility. In this paper, we explore factors that influence radiocarbon dating variabilities. We obtained estimated radiocarbon ages by sending identical samples to several labs multiple times. A Bayesian method was used to analyze the obtained data. From the analysis, we conclude that some factors (such as type of labs and megasamples) can induce variability when estimating radiocarbon age. We identify the size of variability caused by each factor and analyze the estimated variability in each lab corresponds with the reported variability.
Principal Component Analysis with Coefficient of Variation Matrix
Kim, Ji-Hyun ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 385~392
DOI : 10.5351/KJAS.2015.28.3.385
Principal component analysis (PCA), a dimension-reduction technique, is usually implemented after the variables are standardized when the measurement unit of variables are different. To standardize a variable we divide it by its standard deviation. But there is another way to transform a variable to be independent of its measurement unit. It is to divide it by its mean rather than standard deviation. Implementing PCA on standardized variables is equivalent to implementing PCA with a correlation matrix of original variables. Similarly, implementing PCA on the transformed variables divided by their means is equivalent to implementing PCA with a matrix related to the coefficients of variation of the original variables. We explain why we need to implement PCA on the variables transformed by their means.
A Two Factor Model with Mean Reverting Process for Stochastic Mortality
Lee, Kangsoo ; Jho, Jae Hoon ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 393~406
DOI : 10.5351/KJAS.2015.28.3.393
We examine how to model mortality risk using the adaptation of the mean-reverting processes for the two factor model proposed by Cairns et al. (2006b). Mortality improvements have been recently observed in some countries such as United Kingdom; therefore, we assume long-run mortality converges towards a trend at some unknown time and the mean-reverting processes could therefore be an appropriate stochastic model. We estimate the parameters of the two-factor model incorporated with mean-reverting processes by a Metropolis-Hastings algorithm to fit United Kingdom mortality data from 1991 to 2015. We forecast the evolution of the mortality from 2014 to 2040 based on the estimation results in order to evaluate the issue price of a longevity bond of 25 years maturity. As an application, we propose a method to quantify the speed of mortality improvement by the average mean reverting times of the processes.
Categorical Variable Selection in Naïve Bayes Classification
Kim, Min-Sun ; Choi, Hosik ; Park, Changyi ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 407~415
DOI : 10.5351/KJAS.2015.28.3.407
Bayes Classification is based on input variables that are a conditionally independent given output variable. The
Bayes assumption is unrealistic but simplifies the problem of high dimensional joint probability estimation into a series of univariate probability estimations. Thus
Bayes classier is often adopted in the analysis of massive data sets such as in spam e-mail filtering and recommendation systems. In this paper, we propose a variable selection method based on
statistic on input and output variables. The proposed method retains the simplicity of
Bayes classier in terms of data processing and computation; however, it can select relevant variables. It is expected that our method can be useful in classification problems for ultra-high dimensional or big data such as the classification of diseases based on single nucleotide polymorphisms(SNPs).
Usage and Estimation of R-indicator for Representative
Park, Hyeonah ; Lee, Kee-Jae ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 417~427
DOI : 10.5351/KJAS.2015.28.3.417
Measures in response rate used to measure the representativeness of the sample (the more high response rate) better explain the representativeness of the sample. However, we cannot often explain the representativeness of the sample because there is nonresponse even in the high response rate. Therefore, Schouten et al. (2009) presented a new R-indicator measure that can be described as a representative of the sample. We research the new estimator of the R-indicator in this paper because there are parameters that require estimations. We describe the meanings as representative of the R-indicator; consequently, the bias and efficiency of the proposed estimator for R-indicator are compared to the existing estimator under various simulations. The representativeness of the sample is also explained by applying the proposed estimators in the actual data.
Volatility Forecasting of Korea Composite Stock Price Index with MRS-GARCH Model
Huh, Jinyoung ; Seong, Byeongchan ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 429~442
DOI : 10.5351/KJAS.2015.28.3.429
Volatility forecasting in financial markets is an important issue because it is directly related to the profit of return. The volatility is generally modeled as time-varying conditional heteroskedasticity. A generalized autoregressive conditional heteroskedastic (GARCH) model is often used for modeling; however, it is not suitable to reflect structural changes (such as a financial crisis or debt crisis) into the volatility. As a remedy, we introduce the Markov regime switching GARCH (MRS-GARCH) model. For the empirical example, we analyze and forecast the volatility of the daily Korea Composite Stock Price Index (KOSPI) data from January 4, 2000 to October 30, 2014. The result shows that the regime of low volatility persists with a leverage effect. We also observe that the performance of MRS-GARCH is superior to other GARCH models for in-sample fitting; in addition, it is also superior to other models for long-term forecasting in out-of-sample fitting. The MRS-GARCH model can be a good alternative to GARCH-type models because it can reflect financial market structural changes into modeling and volatility forecasting.
Course Probability of Yut according to Starting Order
Cho, Daehyeon ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 443~455
DOI : 10.5351/KJAS.2015.28.3.443
The Korean game of yut is a traditional games that everyone can enjoy regardless of gender or ages. Yut consists of four sticks with a Head and Tail. We are interested in the course probabilities in the game of yut that are different according to the starting order of the four pieces of yut. So we consider the probabilities of five results of yut which we toss according to the probability of Head. We calculate probabilities according to 4 courses where one piece of yut can go through in a yutpan according to the starting order of each piece of yut.
Predictability of Consumer Expectations for Future Changes in Real Growth
Kim, Tae-Ho ; Lim, La-Hee ; Lee, Seung-Eun ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 457~465
DOI : 10.5351/KJAS.2015.28.3.457
The long lasting world-wide recession and low economic progress have made it more important to predict future economic behavior. Accordingly, it is of interest to explore useful leading indicators, correlated with policy targets, to predict future economic growth. This study attempts to develop a model to evaluate the performance of consumer survey results from Statistics Korea to predict future economic activities. A statistical model is formulated and estimated to generate predictions by utilizing consumer expectations. The prediction is found improved in the distant future and consumer expectations appear to be a useful leading indicator to provide information of future real growth.
A Regression based Unconstraining Demand Method in Revenue Management
Lee, JaeJune ; Lee, Woojoo ; Kim, Junghwan ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 467~475
DOI : 10.5351/KJAS.2015.28.3.467
Accurate demand forecasting is a crucial component in revenue management(RM). The booking data of departed flights is used to forecast the demand for future departing flights; however, some booking requests that were denied were omitted in the departed flights data. Denied booking requests can be interpreted as censored in statistics. Thus, unconstraining demand is an important issue to forecast the true demands of future flights. Several unconstraining methods have been introduced and a method based on expectation maximization is considered superior. In this study, we propose a new unconstraining method based on a regression model that can entertain such censored data. Through a simulation study, the performance of the proposed method was evaluated with two representative unconstraining methods widely used in RM.
A Note on the Decision of Sample Size by Relative Standard Error in Successive Occasions
Han, GeunShik ; Lee, Gi-Sung ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 477~483
DOI : 10.5351/KJAS.2015.28.3.477
This study deals with the decision problem of sample size by the relative standard error of estimates derived from survey results in successive occasions. The population of the construction in business survey results is used to calculate quartile of the relative standard error of the 1,000 sample obtained from simple or stratified random sampling. The sample size at time t with a relative standard error of the point (t-1) in the successive occasions were calculated according to the sampling method. As a result, in terms of the sample size according to the size of the relative standard error of the (t-1), simple random sampling differs significantly from stratified sampling. In addition, we could see differences in sample size (depending on how the population is stratified) and that careful attention is required in the problem of sample size by the relative standard error of estimates derived from survey results in successive occasions.
Applications of Bootstrap Methods for Canonical Correspondence Analysis
Ko, Hyeon-Seok ; Jhun, Myoungshic ; Jeong, Hyeong Chul ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 485~494
DOI : 10.5351/KJAS.2015.28.3.485
Canonical correspondence analysis is an ordination method used to visualize the relationships among sites, species and environmental variables. However, projection results are fluctuations if the samples slightly change and consistent interpretation on ecological similarity among species tends to be difficult. We use the bootstrap methods for canonical correspondence analysis to solve this problem. The bootstrap method results show that the variations of coordinate points are inversely proportional to the number of observations and coverage rates with bootstrap confidence interval approximates to nominal probabilities.
Classification Analysis for Unbalanced Data
Kim, Dongah ; Kang, Suyeon ; Song, Jongwoo ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 495~509
DOI : 10.5351/KJAS.2015.28.3.495
We study a classification problem of significant differences in the proportion of two groups known as the unbalanced classification problem. It is usually more difficult to classify classes accurately in unbalanced data than balanced data. Most observations are likely to be classified to the bigger group if we apply classification methods to the unbalanced data because it can minimize the misclassification loss. However, this smaller group is misclassified as the larger group problem that can cause a bigger loss in most real applications. We compare several classification methods for the unbalanced data using sampling techniques (up and down sampling). We also check the total loss of different classification methods when the asymmetric loss is applied to simulated and real data. We use the misclassification rate, G-mean, ROC and AUC (area under the curve) for the performance comparison.
Re-Transformation of Power Transformation for ARMA(p, q) Model - Simulation Study
Kang, Jun-Hoon ; Shin, Key-Il ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 511~527
DOI : 10.5351/KJAS.2015.28.3.511
For time series analysis, power transformation (especially log-transformation) is widely used for variance stabilization or normalization for stationary ARMA(p, q) model. A simple and naive back transformed forecast is obtained by taking the inverse function of expectation. However, this back transformed forecast has a bias. Under the assumption that the log-transformed data is normally distributed. The unbiased back transformed forecast can be obtained by the expectation of log-normal distribution; consequently, the property of this back transformation was studied by Granger and Newbold (1976). We investigate the sensitivity of back transformed forecasts under several different underlying distributions using simulation studies.
Comparison of Single Imputation Methods in 2×2 Cross-Over Design with Missing Observations
Jo, Bobae ; Kim, Dongjae ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 529~540
DOI : 10.5351/KJAS.2015.28.3.529
A cross-over design is frequently used in clinical trials (especially in bioequivalence tests with a parametric method) for the comparison of two treatments. Missing values frequently take place in cross-over designs in the second period. Usually, subjects that have missing values are removed and analyzed. However, it can be unsuitable in clinical trials with a small sample size. In this paper, we compare single imputation methods in a
cross-over design when missing values exist in the second period. Additionally, parametric and nonparametric methods are compared after applying single imputation methods. A Monte-Carlo simulation study compares type I error and the power of methods.
Performance Analysis of Volatility Models for Estimating Portfolio Value at Risk
Yeo, Sung Chil ; Li, Zhaojing ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 541~559
DOI : 10.5351/KJAS.2015.28.3.541
VaR is now widely used as an important tool to evaluate and manage financial risks. In particular, it is important to select an appropriate volatility model for the rate of return of financial assets. In this study, both univariate and multivariate models are considered to evaluate VaR of the portfolio composed of KOSPI, Hang-Seng, Nikkei indexes, and their performances are compared through back testing techniques. Overall, multivariate models are shown to be more appropriate than univariate models to estimate the portfolio VaR, in particular DCC and ADCC models are shown to be more superior than others.
Robust Extrapolation Design Criteria under the Uncertainty of Model and Error Structure
Jang, Dae-Heung ; Kim, Youngil ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 561~571
DOI : 10.5351/KJAS.2015.28.3.561
When we consider an optimal design to predict the response corresponding to the point outside the design region, we are extremely careful about choosing the design criteria for selecting the support points. The assumed model and its accompanying error structure should be assumed to extend beyond the design region for the selected design criteria to be valid. Thus, we modify the existing design criteria such as extrapolation-optimality to be suited to those situations. We propose some maximin approaches in this paper. Simple and quadratic regression models are tested to find the basic characteristics of such maximin approaches. Some main findings are discussed in the conclusion.
Sample Size Determination for One-Sample Location Tests
Yeo, In-Kwon ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 573~581
DOI : 10.5351/KJAS.2015.28.3.573
We study problems of sample size determination for one-sample location tests. A simulation study shows that sample size calculations based on approximated distribution do not achieve the nominal level of power. We investigate sample size determinations based on exact distribution and with a power that attains the nominal level.
Zero-Inflated INGARCH Using Conditional Poisson and Negative Binomial: Data Application
Yoon, J.E. ; Hwang, S.Y. ;
Korean Journal of Applied Statistics, volume 28, issue 3, 2015, Pages 583~592
DOI : 10.5351/KJAS.2015.28.3.583
Zero-inflation has recently attracted much attention in integer-valued time series. This article deals with conditional variance (volatility) modeling for the zero-inflated count time series. We incorporate zero-inflation property into integer-valued GARCH (INGARCH) via conditional Poisson and negative binomial marginals. The Cholera frequency time series is analyzed as a data application. Estimation is carried out using EM-algorithm as suggested by Zhu (2012).