• The Korean Journal of Applied Statistics
• /
• v.27 no.6
• /
• pp.991-1002
• /
• 2014

Both Bayesian and maximum likelihood methods are efficient for the estimation of regression coefficients of various Tobit regression models (see. e.g. Chib, 1992; Greene, 1990; Lee and Choi, 2013); however, some researchers recognized that the maximum likelihood method tends to underestimate the disturbance variance, which has implications for the estimation of marginal effects and the asymptotic standard error of estimates. The underestimation of the maximum likelihood estimate in a seemingly unrelated Tobit regression model is examined. A Bayesian method based on an objective noninformative prior is shown to provide proper estimates of the disturbance variance as well as other regression parameters

• Journal of Animal Science and Technology
• /
• v.57 no.5
• /
• pp.19.1-19.6
• /
• 2015

The objective of this study was to estimate variance components and genetic parameters for growth curve parameters in Guilan sheep. Studied traits were parameters of Brody growth model which included A (asymptotic mature weight), B (initial animal weight) and K (maturation rate). The data set and pedigree information used in this study were obtained from the Agricultural Organization of Guilan province (Rasht, Iran) and comprised 8647 growth curve records of lambs from birth to 240 days of age during 1994 to 2014. Marginal posterior distributions of parameters and variance components were estimated using TM program. The Gibbs sampler was run 300000 rounds and the first 60000 rounds were discarded as a burn-in period. Posterior mean estimates of direct heritabilities for A, B and K were 0.39, 0.23 and 0.039, respectively. Estimates of direct genetic correlation between growth curve parameters were 0.57, 0.03 and -0.01 between A-B, A-K and B-K, respectively. Estimates of direct genetic trends for A, B and K were positive and their corresponding values were $0.014{\pm}0.003$ (P < 0.001), $0.0012{\pm}0.0009$ (P > 0.05) and $0.000002{\pm}0.0001$ (P > 0.05), respectively. Residual correlations between growth curve parameters varied form -0.52 (between A-K) to 0.48 (between A-B). Also, phenotypic correlations between growth curve parameters varied form -0.49 (between A-K) to 0.47 (between A-B). The results of this study indicated that improvement of growth curve parameters of Guilan sheep seems feasible in selection programs. It is worthwhile to develop a selection strategy to obtain an appropriate shape of growth curve through changing genetically the parameters of growth model.

• Journal of the Korean Data and Information Science Society
• /
• v.7 no.1
• /
• pp.47-57
• /
• 1996

This paper on planning constant accelerated life test is assumed that parameters for a lognormal life distribution are depended on changes of stresses. The proposed test plans are optimum in that they minimize the asymptotic variance of maximum likelihood estimator of a specified quantile at the design stress. The optimal amount of low stress level ${\xi}_{L}$ and optimal sample proportion ${\pi}$ to be allocated at low stress level are obtained when the ratio of scales at high stress level and design stress level is unknown.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.1682-1685
• /
• 2005

System algorithms estimated by deterministic input may occur the error between predicted and actual output. Especially, actual system can't predict the exact outputs due to uncertainty and tolerance of input parameters. A single output to a set of inputs has a limited value without the variation. Hence, we should consider various scatters caused by the load assessment, material characteristics, stress analysis and manufacturing methods in order to perform the robust design or estimate the reliability of structure. The system design with uncertainty should perform the probabilistic structural optimization with the statistical response and the reliability. This method calculated the probability distributions of the characteristics such as stress by combining stress analysis, response surface methodology and Monte-Carlo Method and got the probabilistic sensitivity. The sensitivity of structural response with respect to inconstant design variables was estimated by fracture probability. Therefore, this paper proposed the probabilistic reliability design method for fracture of uncorved freight end beam and the design criteria by fracture probability.

• The Korean Journal of Applied Statistics
• /
• v.22 no.5
• /
• pp.1073-1084
• /
• 2009

For ordinal categorical $R{\times}R$ tables, a weighted measure of association, $H_{\omega}$, was proposed and its maximum likelihood estimator and asymptotic variance were drived. We redefined the last paradox of ${\kappa}$ and proved its relation to marginal distributions. We also introduced the new paradox of ${\kappa}$ and summaried the general relationships between ${\kappa}$ and marginal distributions.

• Journal of Korean Society for Quality Management
• /
• v.25 no.1
• /
• pp.80-99
• /
• 1997

This paper considers designing the simple (2-level) constant-and step-stress ALTs minimizing the asymptotic variance of the maximum likelihood estimator of the accelaeration factor, which is defined as the ratio of the 100qth percentile at use stress to that a specified stress, for items having lognormally-distributed lives. It is assumed that (i) the log-linear relationship exists between the stress and the mean log life, (ii) the standard deviation of the log life is constant, and (iii) the cumulative exposure model holds for the effect of changing stress. For the constant-stress ALT the low stress and the sample proportion allocated to low stress are determined and for two modes of stress loading of step-stress ALTs, the low-to-high and high-to-low, the low stress and the stress change time are determined. For selected values of the design parameters the optimum plans are figured, two modes of step-stress ALTs and the constant-stress ALT are compared to each other, and the effects of the incorrect pre-estimates of the design parameters are investigated.

• Communications for Statistical Applications and Methods
• /
• v.16 no.5
• /
• pp.753-762
• /
• 2009

A measure of agreement H in$2{\times}2$ contingency table was proposed by Park and Park (2007) to resolve the two paradoxes of k. In this study, we generalize H to where the number of categories is greater than two and derive its asymptotic large-sample variance. We also explain the relationships between k's paradoxes and marginal distributions. Using some examples of $3{\times}3$ contingency tables, the behaviors of H and other measures of agreement are compared.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.37 no.3
• /
• pp.135-150
• /
• 2012

This paper presents a portfolio model for a long-term power generation mix problem. The proposed portfolio model evaluates generation mix by considering the tradeoffs between the expected cost for power generation and its variability. Unlike conventional portfolio models measuring variance, we introduce Conditional Value-at-Risk (CVaR) in designing the variability with aims to considering events that are enormously expensive but are rare such as nuclear power plant accidents. Further, we consider uncertainties associated with future electricity demand, fuel prices and their correlations, and capital costs for power plant investments. To obtain an objective generation by each energy source, we employ the sample average approximation method that approximates the stochastic objective function by taking the average of large sample values so that provides asymptotic convergence of optimal solutions. In addition, the method includes Monte Carlo simulation techniques in generating random samples from multivariate distributions. Applications of the proposed model and method are demonstrated through a case study of an electricity industry with nuclear, coal, oil (OCGT), and LNG (CCGT) in South Korea.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.1
• /
• pp.151-156
• /
• 2012

In this paper robust nonlinear image denoising algorithms are introduced for the distribution which is Gaussian in the center and Laplacian in the tails. The distribution is known as the least favorable ${\epsilon}$-contaminated normal distribution that maximizes the asymptotic variance. The proposed filter proves to be the maximum likelihood estimator under the heavy-tailed Gaussian noise environments. It is optimal in the respect of maximizing the efficacy under the above noise environment. Another filter for reducing impulsive noise is proposed by mixing with the myriad filter to propose an amplitude-limited myriad filter. Extensive experiment is conducted with images corrupted with ${\alpha}$-stable noise to analyze the behavior and performance of the proposed filters.

• Journal of Communications and Networks
• /
• v.18 no.3
• /
• pp.351-363
• /
• 2016

This paper investigates the performance of interference-limited two-way transmission protocols in the Rayleigh fading channels. New lower bound of outage probability and approximate expression of bit error rate (BER) for three-phase two-way relaying (3P-TWR) protocol are derived in closed-form. These expressions are valid for arbitrary signal-to-noise ratio values, numbers of co-channel interferers and amajority of modulation formats employed in the practical system. Then a comparative study is developed for the performance of three two-way transmission protocols, i.e., direct transmission (DT) protocol, two-phase two-way relaying (2P-TWR) protocol and 3P-TWR protocol based on the asymptotic expressions of outage probability and BER. On the basis of the theoretical results, the thresholds on the strength (variance) of direct channel and target rate for the relative performance of different protocols are obtained and the effect of interferences at the terminal and relay on the relative performance is analyzed. The results present key insights on how to choose proper two-way transmission protocol with the knowledge of fading channels, required date rate and modulation format to optimize the system performance in the practical interference-limited scenarios. Simulation results are presented to validate the theoretical analysis.