• Journal of the Korean Data and Information Science Society
• /
• v.14 no.2
• /
• pp.313-323
• /
• 2003

For the growth curve model with arbitrary covariance structure, known as unstructured covariance matrix, the problems of detecting outliers are discussed in this paper. In order to detect outliers in the growth curve model, the test statistics using U-distribution is established. After detecting outliers in growth curve model, we test homo and/or hetero-geneous covariance matrices using PSR Quasi-Bayes Criterion. For illustration, one numerical example is discussed, which compares between before and after outlier deleting.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.161-171
• /
• 2006

In this paper, we consider the random number generation method for multivariate Poisson distribution with specific covariance matrix. Random number generating method for the multivariate Poisson distribution is considered into two part, by first solving the linear equation to determine the univariate Poisson parameter, then convoluting independent univariate Poisson variates with appropriate expectations. We propose a numerical algorithm to solve the linear equation given the specific covariance matrix.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.449-451
• /
• 2003

The estimation of time-series data is fundamental process in many data analysis cases. However, the unwanted measurement error is usually added to true data, so that the exact estimation depends on efficient method to eliminate the error components. The wavelet transform method nowadays is expected to improve the accuracy of estimation, because it is able to decompose and analyze the data in various resolutions. Therefore, the wavelet based Kalman filter method for the estimation of time-series data is proposed in this paper. The wavelet transform separates the data in accordance with frequency bandwidth, and the detail wavelet coefficient reflects the stochastic process of error components. This property makes it possible to obtain the covariance of measurement error. We attempt the estimation of true data through recursive Kalman filtering algorithm with the obtained covariance value. The procedure is verified with the fundamental example of Brownian walk process.

• The Korean Journal of Applied Statistics
• /
• v.37 no.4
• /
• pp.455 -465
• /
• 2024

SMOTE (synthetic minority over-sampling technique) has been used the most as a solution to the problem of imbalanced data. SMOTE selects the nearest neighbor based on Euclidean distance. However, Euclidean distance has the disadvantage of not considering the correlation between variables. In particular, the Mahalanobis distance has the advantage of considering the covariance of variables. But if there are outliers, they usually influence calculating the Mahalanobis distance. To solve this problem, we use the Mahalanobis distance by estimating the covariance matrix using MCD (minimum covariance determinant). Then apply Mahalanobis distance based on MCD to SMOTE to create new data. Therefore, we showed that in most cases this method provided high performance indicators for classifying imbalanced data.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.755-771
• /
• 2014

Isotropy is one of the main assumptions for the ease of spatial prediction (named kriging) based on some covariance models. A lack of isotropy (or anisotropy) in a spatial process necessitates that some additional parameters (angle and ratio) for anisotropic covariance model be obtained in order to produce a more reliable prediction. In this paper, we propose a new class of geometrically extended anisotropic covariance models expressed as a weighted average of some geometrically anisotropic models. The maximum likelihood estimation method is taken into account to estimate the parameters of our interest. We evaluate the performances of our proposal and compare it with an isotropic covariance model and a geometrically anisotropic model in simulation studies. We also employ extended geometric anisotropy to the analysis of real data.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.3
• /
• pp.983-990
• /
• 2006

In this paper, we study an asymmetric power fractionally integrated GARCH model and find a region on which the process is stationary ergodic and has long memory property.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.539-548
• /
• 2016

This paper is a study on the multivariate CUSUM control charts using three different control statistics for monitoring covariance matrix. We get control limits and ARLs of the proposed multivariate CUSUM control charts using three different control statistics by using computer simulations. The performances of these proposed multivariate CUSUM control charts have been investigated by comparing ARLs. The purpose of control charts is to detect assignable causes of variation so that these causes can be found and eliminated from process, variability will be reduced and the process will be improved. We show that the charts based on three different control statistics are very effective in detecting shifts, especially shifts in covariances when the variables are highly correlated. When variables are highly correlated, our overall recommendation is to use the multivariate CUSUM control charts using trace for detecting changes in covariance matrix.

• Computers and Concrete
• /
• v.13 no.5
• /
• pp.659-677
• /
• 2014

A Bayesian response surface updating procedure is applied in order to update the parameters of the covariance function of a random field for concrete properties based on a limited number of available measurements. Formulas as well as a numerical algorithm are presented in order to update the parameters of response surfaces using Markov Chain Monte Carlo simulations. The parameters of the covariance function are often based on some kind of expert judgment due the lack of sufficient measurement data. However, a Bayesian updating technique enables to estimate the parameters of the covariance function more rigorously and with less ambiguity. Prior information can be incorporated in the form of vague or informative priors. The proposed estimation procedure is evaluated through numerical simulations and compared to the commonly used least square method.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.815-825
• /
• 2016

Marginalized random effects models (MREMs) are often used to analyze longitudinal categorical data. The models permit direct estimation of marginal mean parameters and specify the serial correlation of longitudinal categorical data via the random effects. However, it is not easy to estimate the random effects covariance matrix in the MREMs because the matrix is high-dimensional and must be positive-definite. To solve these restrictions, we introduce two modeling approaches of the random effects covariance matrix: partial autocorrelation and the modified Cholesky decomposition. These proposed methods are illustrated with the real data from Korean genomic epidemiology study.

• Journal of Environmental Science International
• /
• v.22 no.1
• /
• pp.119-129
• /
• 2013

We compared equilibrium evaporation($E_{equili}$) eddy-covariance($E_{eddy}$) with soil moisture data($E_{SMseries}$) which were measured with a 2 hours sampling interval at three points for a humid forest hillslope from May 5th to May 31th in 2009. Accumulations of $E_{eddy}$, $E_{equili}$ for the study period were estimated as 2.52, 3.28 mm and those of $E_{SMseries}$ were ranged from 1.91 to 2.88 mm. It suggested that the eddy-covariance method considering the spatial heterogeneity of soil evaporation is useful to evaluate the soil evaporation. Method A, B and C were proposed using mean meterological data and daily moisture variation and the computations were compared to eddy-covariance method and equilibrium evaporation. The methods using soil moisture data can describe the variations of soil evaporation from eddy-covariance through simple moving average analysis. Method B showed a good matched with eddy-covariance method. This indicated that Dry Surface Layer (DSL) at 14:00 which was used for method B is important variable for the evaluation of soil evaporation. The total equilibrium evaporation was not significantly different to those of the others. However, equilibrium evaporation showed a problem in estimating soil evaporation because the temporal tendency of $E_{equili}$ was not related with the those of the other methods. The improved understanding of the soil evaporation presented in this study will contribute to the understandings of water cycles in a forest hillslope.