• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.268-275
• /
• 2003

We consider the Bayesian accelerated failure time model. The error distribution is assigned a skewed normal distribution which is including normal distribution. For noninformative priors of regression coefficients, we show the propriety of posterior distribution. A Markov Chain Monte Carlo algorithm(i.e., Gibbs Sampler) is used to obtain a predictive distribution for a future observation and Bayes estimates of regression coefficients.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.809-818
• /
• 2014

Multivariate skew-normal distribution(distribution that includes multivariate normal distribution) has been recently applied to many application areas. We consider saddlepoint approximation for a statistic of linear combination based on a multivariate skew-normal distribution. This approach can be regarded as an extension of Na and Yu (2013) that dealt saddlepoint approximation for the distribution of a skew-normal sample mean for a linear statistic and multivariate version. Simulations results and examples with real data verify the accuracy and applicability of suggested approximations.

• Communications for Statistical Applications and Methods
• /
• v.19 no.3
• /
• pp.345-358
• /
• 2012

In a parametric sample selection model, the distribution assumption is critical to obtain consistent estimates. Conventionally, the normality assumption has been adopted for both error terms in selection and main equations of the model. The normality assumption, however, may excessively restrict the true underlying distribution of the model. This study introduces the $S_U$-normal distribution into the error distribution of a sample selection model. The $S_U$-normal distribution can accommodate a wide range of skewness and kurtosis compared to the normal distribution. It also includes the normal distribution as a limiting distribution. Moreover, the $S_U$-normal distribution can be easily extended to multivariate dimensions. We provide the log-likelihood function and expected value formula based on a bivariate $S_U$-normal distribution in a sample selection model. The results of simulations indicate the $S_U$-normal model outperforms the normal model for the consistency of estimators. As an empirical application, we provide the sample selection model for car ownership and a car expense relationship.

• Steel and Composite Structures
• /
• v.44 no.5
• /
• pp.619-632
• /
• 2022

There are numerous structural details (Longitudinal beam, web plate, U-ribs and I-ribs) in the top and bottom plates of steel box girders, which have significant influences on the longitudinal stress (normal stress) distribution. Clarifying the influence of these structural details on the normal stress distribution is important. In this paper, the ultra-wide steel box girder with large cantilevers of the Jinhai Bridge in China, which is the widest cable-stayed bridge in the world, has been analyzed. A 1:4.5 scale laboratory model of the steel box girder has been manufactured, and the influence of structural details on the normal stress distribution in the top and bottom plates for four different load cases has been analyzed in detail. Furthermore, a three-dimensional finite element model has been established to further investigate the influence regularity of structural details on the normal stress. The experimental and finite element analysis (FEA) results have shown that different structural details of the top and bottom plates have varying effects on the normal stress distribution. Notably, the U-ribs and I-ribs of the top and bottom plates introduce periodicity to the normal stress distribution. The period of the influence of U-ribs on the normal stress distribution is the sum of the single U-rib width and the U-rib spacing, and that of the influence of I-ribs on the normal stress distribution is equal to the spacing of the I-ribs. Furthermore, the same structural details but located at different positions, will have a different effect on the normal stress distribution.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.2
• /
• pp.97-103
• /
• 2006

Recently, the control chart is developed for monitoring processes with normal short production runs by the coefficient of variation(CV) characteristic for a normal distribution. This control chart does not work well in non-normal short production runs. And most of industrial processes are known to follow the non-normal distribution. Therefore, the control chart is required to be developed for monitoring the processes with non-normal short production runs by the CV characteristics for a non-normal distribution. In this paper, we suggest the control chart for monitoring the processes with a gamma short runs by the CV characteristics for a gamma distribution. This control chart is denoted by the gamma CV control chart. Futhermore evaluated the performance of the gamma CV control chart by average run length(ARL).

• Communications for Statistical Applications and Methods
• /
• v.15 no.3
• /
• pp.325-332
• /
• 2008

We consider estimation of reliability P(Y < X) and distribution of the ratio when X and Y are independent uniform random variable and power function random variable, respectively and also consider the estimation problem when X and Y are independent uniform random variable and a half-normal random variable, respectively.

• Journal of Korean Society for Quality Management
• /
• v.44 no.1
• /
• pp.167-180
• /
• 2016

Purpose: The problem for the traditional control chart is that it is unable to monitor the very small fraction of nonconforming and the underlying distribution is the normal distribution. $Z_p$ control chart is useful where it controls the vert small fraction on nonconforming. In this study, we will design the $Z_p$ control chart in order to use under non-normal process. Methods: $Z_p$ is calculated not by failure rate based on attribute data but using variable data. Control limit for non-normal $Z_p$ control chart is designed based on ${\alpha}$-risk calculated by cumulative distribution function of Burr distribution. ${\beta}$-risk, which is for performance evaluation, obtains in the Burr distribution's cumulative distribution function and control limit. Results: The control limit for non-normal $Z_p$ control chart is designed based on Burr distribution. The sensitivity can be checked through ARL table and OC curve. Conclusion: Non-normal $Z_p$ control chart is able to control not only the very small fraction of nonconforming, but it is also useful when $Z_p$ distribution is non-normal distribution.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.571-579
• /
• 2016

Most of studies related to the distributions of quadratic forms are conducted under the assumption of multivariate normal distribution. In this paper, we suggested an approximation to the distribution of quadratic forms based on multivariate skew-normal distribution as alternatives for multivariate normal distribution. Saddlepoint approximations are considered and the accuracy of the approximations are verified through simulation studies.

• Journal of the Korean Ceramic Society
• /
• v.34 no.9
• /
• pp.927-934
• /
• 1997

Static and cyclic fatigue tests were carried out for alumina specimen to study the statistical analyses (normal, lognormal and Weibull distribution) of fatigue lifetime data and nominal initial crack length data. Fatigue lifetime data followed Weibull distribution better than normal or lognormal distribution, for the shape parameter of the notched specimen was larger than that of the unnotched specimen. The nominal initial crack length data obtained from fatigue lifetime followed the lognormal and Weibull distribution better than normal distribution, for the coefficient of variation of the unnotched specimen was larger than that of the notched specimen, and shape parameter of unnotched specimen was smaller than that of the notched specimen.

• Journal of applied mathematics & informatics
• /
• v.27 no.5_6
• /
• pp.1429-1433
• /
• 2009

The likelihood distance for the joint distribution of two multivariate normal distributions with common covariance matrix is explicitly derived. It is useful for identifying outliers which do not follow the joint multivariate normal distribution with common covariance matrix. The likelihood distance derived here is a good ground for the use of a generalized Wilks statistic in influence analysis of two multivariate normal data.