• Communications for Statistical Applications and Methods
• /
• v.21 no.6
• /
• pp.539-559
• /
• 2014

We derive the exact distribution of summation for random samples from uniform distribution and then compare the exact distribution with the approximated normal distribution obtained by the central limit theorem. To check the similarity between two distributions, we consider five existing normality tests based on the difference between the target normal distribution and empirical distribution: Anderson-Darling test, Kolmogorov-Smirnov test, Cramer-von Mises test, Shapiro-Wilk test and Shaprio-Francia test. For the purpose of comparison, those normality tests are applied to the simulated data. It can sometimes be difficult to derive an exact distribution. Thus, we try two different transformations to find out which transform is easier to get the exact distribution in terms of calculation complexity. We compare two transformations and comment on the advantages and disadvantages for each transformation.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.277-289
• /
• 2003

We propose modified exact inferential methods in logistic regression model. Exact conditional distribution in logistic regression model is often highly discrete, and ordinary exact inference in logistic regression is conservative, because of the discreteness of the distribution. For the exact inference in logistic regression model we utilize the modified P-value. The modified P-value can not exceed the ordinary P-value, so the test of size $\alpha$ based on the modified P-value is less conservative. The modified exact confidence interval maintains at least a fixed confidence level but tends to be much narrower. The approach inverts results of a test with a modified P-value utilizing the test statistic and table probabilities in logistic regression model.

• The Korean Journal of Applied Statistics
• /
• v.15 no.1
• /
• pp.187-199
• /
• 2002

When the sample size is small, exact tests are often employed because the asymptotic distribution of the test statistic is in doubt. The advantage of exact tests is that it is guaranteed to bound the type I error probability to the nominal level. In this paper we review the methods of constructing exact tests, the algorithm and commercial software. We also examine the difference between exact p-values obtained from exact tests and true p-values obtained from the true underlying distribution.

• ETRI Journal
• /
• v.15 no.3
• /
• pp.35-45
• /
• 1994

In this paper we develop an approximation formalism on the queue length distribution for general queueing models. Our formalism is based on two steps of approximation; the first step is to find a lower bound on the exact formula, and subsequently the Chernoff upper bound technique is applied to this lower bound. We demonstrate that for the M/M/1 model our formula is equivalent to the exact solution. For the D/M/1 queue, we find an extremely tight lower bound below the exact formula. On the other hand, our approach shows a tight upper bound on the exact distribution for both the ND/D/1 and M/D/1 queues. We also consider the $M+{\Sigma}N_jD/D/1$ queue and compare our formula with other formalisms for the $M+{\Sigma}N_jD/D/1$ and M+D/D/1 queues.

• The Korean Journal of Applied Statistics
• /
• v.27 no.5
• /
• pp.819-831
• /
• 2014

The Exact-EM algorithm can conventionally fit a mixture of multivariate skew distribution. However, it suffers from highly expensive computational costs to calculate the moments of multivariate truncated t-distribution in E-step. This paper proposes a new SPU-EM method that adopts the AECM algorithm principle proposed by Meng and van Dyk (1997)'s to circumvent the multi-dimensionality of the moments. This method offers a shorter execution time than a conventional Exact-EM algorithm. Some experments are provided to show its effectiveness.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.6
• /
• pp.1109-1115
• /
• 2010

In this paper, exact confidence interval of hyper-geometric parameter, that is the probability of success p in the population is discussed. Usually, binomial distribution is a well known discrete distribution with abundant usage. Hypergeometric distribution frequently replaces a binomial distribution when it is desirable to make allowance for the finiteness of the population size. For example, an application of the hypergeometric distribution arises in describing a probability model for the number of children attacked by an infectious disease, when a fixed number of them are exposed to it. Exact confidence interval estimation of hypergeometric parameter is reviewed. We consider the performance of exact confidence interval estimates of hypergeometric parameter in terms of actual coverage probability by small sample Monte Carlo simulation.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.6
• /
• pp.1175-1182
• /
• 2011

In this paper, properties of exact confidence interval and some approximate confidence intervals of hyper-geometric parameter, that is the probability of success p in the population is discussed. Usually, binomial distribution is a well known discrete distribution with abundant usage. Hypergeometric distribution frequently replaces a binomial distribution when it is desirable to make allowance for the finiteness of the population size. For example, an application of the hypergeometric distribution arises in describing a probability model for the number of children attacked by an infectious disease, when a fixed number of them are exposed to it. Exact confidence interval estimation of hypergeometric parameter is reviewed. We consider the approximation of hypergeometirc distribution to the binomial and normal distribution respectively. Approximate confidence intervals based on these approximation are also adequately discussed. The performance of exact confidence interval estimates and approximate confidence intervals of hypergeometric parameter is compared in terms of actual coverage probability by small sample Monte Carlo simulation.

• Advances in nano research
• /
• v.8 no.2
• /
• pp.135-148
• /
• 2020

The incorporation of carbon nanotubes in a polymer matrix makes it possible to obtain nanocomposite materials with exceptional properties. It's in this scientific background that this work was based. There are several theories that deal with the behavior of plates, in this research based on the Mindlin-Reissner theory that takes into account the transversal shear effect, for analysis of the critical buckling load of a reinforced polymer plate with parabolic distribution of carbon nanotubes. The equations of the model are derived and the critical loads of linear and parabolic distribution of carbon nanotubes are obtained. With different disposition of nanotubes of carbon in the polymer matrix, the effects of different parameters such as the volume fractions, the plate geometric ratios and the number of modes on the critical load buckling are analysed and discussed. The results show that the critical buckling load of parabolic distribution is larger than the linear distribution. This variation is attributed to the concentration of reinforcement (CNTs) at the top and bottom faces for the X-CNT type which make the plate more rigid against buckling.

• Wind and Structures
• /
• v.20 no.4
• /
• pp.489-508
• /
• 2015

In this paper, a novel methodology is proposed to obtain optimum location of outriggers. The method utilizes genetic algorithm (GA) for shape and size optimization of outrigger-braced tall structures. In spite of previous studies (simplified methods), current study is based on exact modeling of the structure in a computer program developed on Matlab in conjunction with OpenSees. In addition to that, exact wind loading distribution is calculated in accordance with ASCE 7-10. This is novel since in previous studies wind loading distributions were assumed to be uniform or triangular. Also, a new penalty coefficient is proposed which is suitable for optimization of tall buildings. Newly proposed penalty coefficient improves the performance of GA and results in a faster convergence. Optimum location and number of outriggers is investigated. Also, contribution of factors like central core and outrigger rigidity is assessed by analyzing several design examples. According to the results of analysis, exact wind load distribution and modeling of all structural elements, yields optimum designs which are in contrast of simplified methods results. For taller frames significant increase of wind pressure changes the optimum location of outriggers obtained by simplified methods. Ratio of optimum location to the height of the structure for minimizing weight and satisfying serviceability constraints is not a fixed value. Ratio highly depends on height of the structure, core and outriggers stiffness and lateral wind loading distribution.

• Journal of the Korean Statistical Society
• /
• v.29 no.4
• /
• pp.489-499
• /
• 2000

We propose a new instrumental variable estimator of the complex parameter of a class of univariate complex-valued diffusion processes defined by the possibly non-stationary and/or nonlinear stochastic differential equations. On the basis of the exact finite sample distribution of the pivotal quantity, we construct the exact confidence intervals and the exact tests for the parameter. Monte-Carlo simulation suggests that the new estimator seems to provide a viable alternative to the maximum likelihood estimator (MLE) for nonlinear and/or non-stationary processes.