• Journal of the Korean Data and Information Science Society
• /
• 제12권1호
• /
• pp.103-125
• /
• 2001

The Box-Cox transformation is a well known family of power transformation that brings a set of data into agreement with the normality assumption of the residuals and hence the response variable of a postulated model in regression analysis. This paper proposes a new method for estimating the Box-Cox transformation using maximization of the Sample Entropy statistic which forces the data to get closer to normal as much as possible. A comparative study of the proposed procedure with the maximum likelihood procedure, the procedure via artificial regression estimation, and the recently introduced maximization of the Shapiro-Francia W' statistic procedure is given. In addition, we generate a table for the optimal spacings parameter in computing the Sample Entropy statistic.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.563-568
• /
• 1995

It is well-known that the sample mean and the sample variance are jointly sufficient under normality assumption. In this paper a proof of the joint sufficiency is given without using the factorization criterion. It is related to a finite Sanov-type conditional theorem, i.e., the conditional probability density of $Y_1$ given sample mean $\mu$ and sample variance $\sigma^2$, where $Y_1, Y_2, \cdots, Y_n$ are independently and identically distributed (i.i.d.) normal random variables with mean m and variance $\delta^2$, equals that of $Y_1$ given sample mean $\mu$ and sample variance $\sigma^2$, where $Y_1, Y_2, \cdots, Y_n$ are i.i.d. normal random variables with mean $\mu$ and variance $\sigma^2$.

• Communications for Statistical Applications and Methods
• /
• 제25권4호
• /
• pp.329-339
• /
• 2018

In this study, we propose tests for equality of several variances with the normality assumption. First of all, we propose the likelihood ratio test by applying the permutation principle. Then by using the p-values for the pairwise tests between variances and combination functions, we propose combination tests. We apply the permutation principle to obtain the overall p-values. Also we review the well- known test statistics for the completion of our discussion and modify a statistic with the p-values. Then we illustrate proposed tests by numerical and simulated data and compare their efficiency with the reviewed ones through a simulation study by obtaining empirical p-values. Finally, we discuss some interesting features related to the resampling methods and tests for equality among several variances.

• 산업경영시스템학회지
• /
• 제25권2호
• /
• pp.50-56
• /
• 2002

In this study, we establish bootstrap control limits for EWMA chart by applying the bootstrap method, called resampling, which could not demand assumptions about pre-distribution when the process is skewed and/or the normality assumption is doubt. The results obtained in this study are summarized as follows : bootstrap EWMA control chart is developed for applying bootstrap method to EWMA chart, which is more sensitive to small shifts of process. With the purpose of eliminating a skewness of the resampling distribution, the bootstrap control limits are established by using a modified residual, and its performance is analyzed by ARL. It is shown that the bootstrap EWMA control chart developed in this study includes the properties of standard EWMA control chart that is sensitive to a small shift, and detects process in out of control more quickly than standard EWMA chart.

• Journal of the Korean Data and Information Science Society
• /
• 제25권3호
• /
• pp.685-696
• /
• 2014

In this paper, we develop Bayesian inference of the finite population mean with the assumption of posterior linearity rather than normality of the superpopulation in the presence of auxiliary information under the balanced loss function. We compare the performance of the optimal Bayes estimator under the balanced loss function with ones of the classical ratio estimator and the usual Bayes estimator in terms of the posterior expected losses, risks and Bayes risks.

• 산업경영시스템학회지
• /
• 제26권3호
• /
• pp.39-49
• /
• 2003

Statistical process cintrol is intended to assist operators of a stable system in monitoring whether a change has occurred in the process, and it uses several control charts as main tools. In design and use of control chart, it is rational that probability of false alarm is minimized in stable process and probability of detecting shifts is maximized in out-of-control. In this study, we establish bootstrap control limits for robust M-estimator chart by applying the bootstrap method, called resampling, which could not demand assumptions about pre-distribution when the process is skewed and/or the normality assumption is doubt. The results obtained in this study are summarized as follows : bootstrap M-estimator control chart is developed for applying bootstrap method to M-estimator chart, which is more robust to keep ARL when process contain contaminate quality characteristic.

• Communications for Statistical Applications and Methods
• /
• 제25권6호
• /
• pp.633-645
• /
• 2018

Gaussian error distributions are a common choice in traditional regression models for the maximum likelihood (ML) method. However, this distributional assumption is often suspicious especially when the error distribution is skewed or has heavy tails. In both cases, the ML method under normality could break down or lose efficiency. In this paper, we consider the log-concave and Gaussian scale mixture distributions for error distributions. For the log-concave errors, we propose to use a smoothed maximum likelihood estimator for stable and faster computation. Based on this, we perform comparative simulation studies to see the performance of coefficient estimates under normal, Gaussian scale mixture, and log-concave errors. In addition, we also consider real data analysis using Stack loss plant data and Korean labor and income panel data.

• Communications for Statistical Applications and Methods
• /
• 제26권1호
• /
• pp.35-46
• /
• 2019

Multivariate Confidence Region (MCR) cannot be used to obtain the confidence region of the mean vector of multivariate data when the normality assumption is not satisfied; however, the Quantile Confidence Region (QCR) could be used with a Multivariate Quantile Vector in these cases. The coverage rate of the QCR is better than MCR; however, it has a disadvantage because the QCR has a wide shape when the probability density function follows a bimodal form. In this study, we propose a Quantile Confidence Region using the Highest density (QCRHD) method with the Highest Density Region (HDR). The coverage rate of QCRHD was superior to MCR, but is found to be similar to QCR. The QCRHD is constructed as one region similar to QCR when the distance of the mean vector is close. When the distance of the mean vector is far, the QCR has one wide region, but the QCRHD has two smaller regions. Based on these features, it is found that the QCRHD can overcome the disadvantages of the QCR, which may have a wide shape.

• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.159-164
• /
• 2019

The statistical inference of Cronbach's alpha measure of internal reliability is known to be inaccurate when sample size is small and the assumption of normality is violated. In this paper, we describe the permutation method in which we compute resampling p-values for testing the difference between two or more independent Cronbach's alpha coefficients. When the over-all permutation test is significant, we also make pairwise post-hoc comparisons using permutaion method. The permutation tests for the equality of two independent Cronbach's alpha coefficients and three independent Cronbach's alpha coefficients are illustrated with an example analysis of survey data.

• 한국산업보건학회지
• /
• 제33권2호
• /
• pp.156-170
• /
• 2023

Objectives: This study conducted practical training to improve the proper usage of personal protective equipment(PPE), which greatly impacts workplace safety and health management. Personal protective equipment education was conducted through active participation, without theoretical modules, and aimed to identify the effects of repeated practical education and determine ways to increase participant satisfaction. Methods: Study data were analyzed using the IBM SPSS Statistics ver.29 software. First, participants' general characteristics were analyzed with frequency analysis. Second, the normality and equality of variances (Leven's test) were tested for the dependent variables prior to statistical analyses to determine the use of parametric tests. In general, normality is assumed when the sample size is 30 or more per the central limit theorem (Park et al., 2014). As our sample size of health management workers was 43, normality can be assumed. However, to ensure rigor of the study, we examined skewness and kurtosis. The results confirmed that the data were normally distributed. Third, the effects of repeated PPE training were analyzed using paired t-tests. Fourth, differences in satisfaction with PPE training according to the safety and health job position and safety and health certification were analyzed with t-test and Welch's t-test. For parameters that did not meet the assumption of equal variances, the Welch's t-test was performed. Results: Repeated PPE training improved the educational outcomes, and the improvements were significant in the 1st and 2nd respiratory PPE and safety and hygiene PPE training evaluations (p<.001). In terms of safety and health job position, repeated training led to improvements in educational outcomes, with significant improvements observed among supervisors and specialized health management institution workers in the 1st and 2nd training evaluations (p<.005). In terms of safety certification, repeated training led to improvements in educational outcomes, with significant improvements observed among both certified and non-certified individuals (p<.005). Regarding satisfaction with PPE training according to safety and health job positions, specialized health management institution workers showed greater satisfaction than supervisors, with significant differences in the satisfaction for expertise of lecture, work relevance, and lecturer's attitude (p<.001). Regarding satisfaction with PPE training according to safety and health certification, satisfaction was higher among certified individuals, with significant differences in satisfaction for work relevance and lecture attitude (p<.05) Conclusions: PPE education should be recommended to be provided as practical training. Repeated training can enhance educational outcomes for individuals with inadequate knowledge and understanding of PPE prior to education. For individuals with high levels of pre-existing knowledge and understanding of PPE, the results show that various training experiences should be provided to enhance their satisfaction. Therefore, it suggests that the workplace should actively seek educational media and methods to acquire expertise and skills in wearing personal protective equipment and improve the ability to use