• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.3
• /
• pp.166-175
• /
• 1998

When we make a speaker adapted model using MAPE (maximum a posteriori estimation), the adapted model has one mixture in each state. This is because we cannot estimate a number of a priori distribution from a speaker-independent model in each state. If the model is represented by one mixture in each state, it is not well adadpted to specific speaker because it is difficult to represent various speech informationof the speaker with one mixture. In this paper, we suggest the method using several mixtures to well represent various speech information of the speaker in each state. But, because speaker-specific training dat is not sufficient, this method can't be used in every state. So, we make the number of mixtures in each state variable in proportion to the number of frames and to the determinant ofthe variance matrix in the state. Using the proposed method, we reduced the error rate than methods using one branch in each state.

• ETRI Journal
• /
• v.34 no.5
• /
• pp.684-689
• /
• 2012

A mode-singular-value-decomposition (SVD) maximum likelihood (ML) estimation procedure is proposed for the source localization problem under an additive measurement error model. In a practical situation, the noise variance is usually unknown. In this paper, we propose an algorithm that does not require the noise covariance matrix as a priori knowledge. In the proposed method, the weight is derived by the inverse of the noise magnitude square in the ML criterion. The performance of the proposed method outperforms that of the existing methods and approximates the Taylor-series ML and Cram$\acute{e}$r-Rao lower bound.

• Nuclear Engineering and Technology
• /
• v.44 no.2
• /
• pp.161-176
• /
• 2012

McCARD is a Monte Carlo (MC) neutron-photon transport simulation code. It has been developed exclusively for the neutronics design of nuclear reactors and fuel systems. It is capable of performing the whole-core neutronics calculations, the reactor fuel burnup analysis, the few group diffusion theory constant generation, sensitivity and uncertainty (S/U) analysis, and uncertainty propagation analysis. It has some special features such as the anterior convergence diagnostics, real variance estimation, neutronics analysis with temperature feedback, $B_1$ theory-augmented few group constants generation, kinetics parameter generation and MC S/U analysis based on the use of adjoint flux. This paper describes the theoretical basis of these features and validation calculations for both neutronics benchmark problems and commercial PWR reactors in operation.

• Journal of Korean Society for Quality Management
• /
• v.22 no.2
• /
• pp.69-78
• /
• 1994

We obtain the Bayes estimator(BE), the minimum variance unbiased estimator(MVUE) and maximun likelihood estimator(MLE) of the reliability when the distribution of the stress and the strength are Weibull with known shape parameters. The experiment is terminated before all of the items on the test have failed and the failed items are partially replaced. Performance of the three estimators for moderate size samples are compared through Monte Carlo simulation.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.249-261
• /
• 2010

Long-term forecasting of seasonal time series is critical in many applications such as planning business strategies and resolving possible problems of a business company. Unlike the traditional approach that depends solely on dynamic models, Li and Hinich (2002) introduced a combination of stochastic dynamic modeling with filter bank approach for forecasting seasonal patterns using highly coherent(High-C) waveforms. We modify the filter selection and forecasting procedure on wavelet domain to be more feasible and compare the resulting predictor with one that obtained from the wavelet variance estimation method. An improvement over other seasonal pattern extraction and forecasting methods based on such as wavelet scalogram, Holt-Winters, and seasonal autoregressive integrated moving average(SARIMA) is shown in terms of the prediction error. The performance of the proposed method is illustrated by a simulation study and an application to the real stock price data.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1455-1464
• /
• 2013

This paper develops maximum likelihood estimators (MLEs) of unknown parameters in an exponentiated half-logistic distribution based on a progressively type-II censored sample. We obtain approximate confidence intervals for the MLEs by using asymptotic variance and covariance matrices. Using importance sampling, we obtain Bayes estimators and corresponding credible intervals with the highest posterior density and Bayes predictive intervals for unknown parameters based on progressively type-II censored data from an exponentiated half logistic distribution. For illustration purposes, we examine the validity of the proposed estimation method by using real and simulated data.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1557-1564
• /
• 2017

We take genomic sequences of high-dimensional low sample size (HDLSS) without ordering of response categories into account. When constructing an appropriate test statistics in this model, the classical multivariate analysis of variance (MANOVA) approach might not be useful owing to very large number of parameters and very small sample size. For these reasons, we present a pseudo marginal model based upon the Gini-Simpson index estimated via Bayesian approach. In view of small sample size, we consider the permutation distribution by every possible n! (equally likely) permutation of the joined sample observations across G groups of (sizes $n_1,{\ldots}n_G$). We simulate data and apply false discovery rate (FDR) and positive false discovery rate (pFDR) with associated proposed test statistics to the data. And we also analyze real SARS data and compute FDR and pFDR. FDR and pFDR procedure along with the associated test statistics for each gene control the FDR and pFDR respectively at any level ${\alpha}$ for the set of p-values by using the exact conditional permutation theory.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.2
• /
• pp.136-144
• /
• 1999

This paper deals with the fault detection problem in uncertain linear multivariable systems and its application. A robust fault detection method presented by Kim et a. (1998) for MIMO (Multi Input/Multi Output) systems has been adopted and applied to the twin rotor MIMO experimental setup using industrial DSP. The system identification problem is formulated for the twin rotor MIMO system and its parameters are estimated using experimental data. Based on the estimated parameters, some fault detection simulations are performed using the robust fault detection method, which shows that the preformance is satisfied.

• Journal of Korean Society for Quality Management
• /
• v.19 no.1
• /
• pp.163-169
• /
• 1991

This paper studies computer programming for Two-Way Layout with Multiple Observations - Fixed Model using the subroutines of the former paper. The following items are considered in the PC computer programming: * significant digits in the computation of Sum of Squares * containing the necessary F-distribution values in the program * including the necessary estimation after the Analysis of Variance * following the rules of KS A 0021 in rounding off digits etc. The running results of Analysis of Variation Table and Estimations of a fictitious example is added with the parts of PC program. It should be mentioned that the main purpose of this paper is in the arousing of the discussion about significant digits concept in the PC computer programming for various kinds of Statistical Methods.

• The Korean Journal of Applied Statistics
• /
• v.5 no.1
• /
• pp.19-28
• /
• 1992

In this paper, minimal balanced $2^t$ fractional factorial designs which permit the estimation of main effects and 2-factor interactions are developed by using a partially balanced array. Such designs are characterized by a minimum number of runs and some balancedness property of the variance-covariance matrix of the estimates. In addition to describing the designs, optimality criteria are discussed and the trace-optimal designs are presented. The proposed designs are especially useful in Taguchi method, where we need to investigate up to 2-factor interactions of the control factors.