• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.6
• /
• pp.391-398
• /
• 2023

In this paper, a penalized maximum likelihood estimation (PMLE) method that applies a penalty to increase the accuracy of a basis-screening-based Kriging model (BSKM) is introduced. The maximum order and set of basis functions used in the BSKM are determined according to their importance. In this regard, the cross-validation error (CVE) for the basis functions is employed as an indicator of importance. When constructing the Kriging model (KM), the maximum order of basis functions is determined, the importance of each basis function is evaluated according to the corresponding maximum order, and finally the optimal set of basis functions is determined. This optimal set is created by adding basis functions one by one in order of importance until the CVE of the KM is minimized. In this process, the KM must be generated repeatedly. Simultaneously, hyper-parameters representing correlations between datasets must be calculated through the maximum likelihood evaluation method. Given that the optimal set of basis functions depends on such hyper-parameters, it has a significant impact on the accuracy of the KM. The PMLE method is applied to accurately calculate hyper-parameters. It was confirmed that the accuracy of a BSKM can be improved by applying it to Branin-Hoo problem.

• Communications for Statistical Applications and Methods
• /
• v.24 no.5
• /
• pp.493-505
• /
• 2017

Maximum likelihood estimation (MLE) of the generalized extreme value distribution (GEVD) is known to sometimes over-estimate the positive value of the shape parameter for the small sample size. The maximum penalized likelihood estimation (MPLE) with Beta penalty function was proposed by some researchers to overcome this problem. But the determination of the hyperparameters (HP) in Beta penalty function is still an issue. This paper presents some data adaptive methods to select the HP of Beta penalty function in the MPLE framework. The idea is to let the data tell us what HP to use. For given data, the optimal HP is obtained from the minimum distance between the MLE and MPLE. A bootstrap-based method is also proposed. These methods are compared with existing approaches. The performance evaluation experiments for GEVD by Monte Carlo simulation show that the proposed methods work well for bias and mean squared error. The methods are applied to Blackstone river data and Korean heavy rainfall data to show better performance over MLE, the method of L-moments estimator, and existing MPLEs.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9C
• /
• pp.734-742
• /
• 2008

In spatially multiplexed MIMO systems that enable high data rate transmission over wireless communication channels, the spatial demultiplexing at the receiver is a challenging task and various demultiplexing methods have been developed. Among the previous methods, maximum likelihood detection with QR decomposition and M-algorithm (QRM-MLD), sphere decoding (SD), QOC, and MOC schemes have been reported to achieve a (near) maximum likelihood (ML) hard decision performance. In general, however, the reliability of soft output of these schemes is not satisfactory. In this paper, we propose a method which enhances the reliability of soft output. By computer simulations, we demonstrate the improved performance by the proposed method.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.39 no.3
• /
• pp.83-95
• /
• 1997

This study was conducted to derive optimal design floods by Gamma distribution models of the annual maximum series at eight watersheds along Geum , Yeong San and Seom Jin river Systems, Design floods obtained by different methods for evaluation of parameters and for plotting positions in the Gamma distribution models were compared by the relative mean errors and graphical fit along with 95% confidence interval plotted on Gamma probability paper. The results were analyzed and summarized as follows. 1.Adequacy for the analysis of flood flow data used in this study was confirmed by the tests of Independence, Homogeneity and detection of Outliers. 2.Basic statistics and parameters were calculated by Gamma distribution models using Methods of Moments and Maximum Likelihood. 3.It was found that design floods derived by the method of maximum likelihood and Hazen plotting position formular of two parameter Gamma distribution are much closer to those of the observed data in comparison with those obtained by other methods for parameters and for plotting positions from the viewpoint of relative mean errors. 4.Reliability of derived design floods by both maximum likelihood and method of moments with two parameter Gamma distribution was acknowledged within 95% confidence interval.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.1
• /
• pp.76-81
• /
• 2010

In this paper, we propose a novel approach to improve the performance of a statistical model-based voice activity detection (VAD) which is based on the second-order conditional maximum a posteriori (CMAP). In our approach, the VAD decision rule is expressed as the geometric mean of likelihood ratios (LRs) based on adapted threshold according to the speech presence probability conditioned on both the current observation and the speech activity decisions in the pervious two frames. Experimental results show that the proposed approach yields better results compared to the statistical model-based and the CMAP-based VAD using the LR test.

• International Journal of Reliability and Applications
• /
• v.6 no.2
• /
• pp.117-133
• /
• 2005

Estimations of the parameters included in a two-component system are derived based on masked system life test data, when the probability of masking depends upon the exact cause of system failure. Also estimations of reliability for the individual components at a specified mission time are derived. Maximum likelihood and Bayes methods are used to derive these estimators. The problem is explained on a series system consisting of two independent components each of which has a Pareto distributed lifetime. Further we present numerical studies using simulation.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.4
• /
• pp.719-731
• /
• 2009

This paper deals with properties of the exponentiated extreme value distribution. We derive the approximate maximum likelihood estimators of the scale parameter and location parameter of the exponentiated extreme value distribution based on multiply Type-II censored samples. We compare the proposed estimators in the sense of the mean squared error for various censored samples.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.619-626
• /
• 2003

Given a prior guess for the unknown value of P(Y

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.603-606
• /
• 2003

The maximum likelihood estimators of the eigenvalues and eigenvectors of $$\Sigma$$_1$^{-1}$$\Sigma$$_2$are shown to be the eigenvalues and eigenvectors of $S$_1$^{1}$S$_2$ under multivariate normality and are explicitly derived. The nature of the eigenvalues and eigenvectors of $$\Sigma$$_1$^{-1}$$\Sigma$$_2$ or their estimators will be uncovered.

• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.639-647
• /
• 2002

Efforts to obtain more power for unit root tests have continued. Pantula at el.(1994) compared empirical powers of several unit root test statistics and addressed that the weighted symmetric estimator(WSE) and the unconditional maximum likelihood estimator(UMLE) are the best among them. One can easily see that the powers of these two statistics are almost the same. In this paper we explain a connection between WSE and UMLE and suggest a unit root test statistic which may explain the connection between them.