• Communications for Statistical Applications and Methods
• /
• 제13권2호
• /
• pp.219-231
• /
• 2006

When the correlation of two random variables is weak, the value of one variable can not be used effectively to predict the other. Even when most of the values are overlapped, it is difficult to find a linear relationship. In this paper, we propose two graphical methods of representing the measures of correlation and independence between two random variables. The first method is used to represent their degree of correlation, and the other is used to represent their independence. Both of these methods are based on the cumulative distribution functions defined in this work.

• Journal of the Korean Statistical Society
• /
• 제22권1호
• /
• pp.13-25
• /
• 1993

The purpose of this paper is to investigate the properties of order statistics under various stochastic relations. We study the stochastic comparison of order statistics in a single sample. And we consider two sample case too. For example, F(t) > G9t) for t > 0 when X and Y are random variables symmetric about 0, with c.d.f.s F and G. Two examples are provided.

• Journal of applied mathematics & informatics
• /
• 제33권5_6호
• /
• pp.517-530
• /
• 2015

In this paper, we establish a class of strong limit theorems, represented by inequalities, for the arbitrary random field with respect to the product binomial distributions indexed by the infinite tree on the generalized random selection system by constructing the consistent distri-bution and a nonnegative martingale with pure analytical methods. As corollaries, some limit properties for the Markov chain field with respect to the binomial distributions indexed by the infinite tree on the generalized random selection system are studied.

• 대한수학회지
• /
• 제53권1호
• /
• pp.45-55
• /
• 2016

Let {$X_n,n{\geq}1$} be a sequence of negatively superadditive dependent random variables. In the paper, we study the strong law of large numbers for general weighted sums ${\frac{1}{g(n)}}{\sum_{i=1}^{n}}{\frac{X_i}{h(i)}}$ of negatively superadditive dependent random variables with non-identical distribution. Some sufficient conditions for the strong law of large numbers are provided. As applications, the Kolmogorov strong law of large numbers and Marcinkiewicz-Zygmund strong law of large numbers for negatively superadditive dependent random variables are obtained. Our results generalize the corresponding ones for independent random variables and negatively associated random variables.

• Communications for Statistical Applications and Methods
• /
• 제27권3호
• /
• pp.285-299
• /
• 2020

In many cases, we are interested in identifying independence between variables. For continuous random variables, correlation coefficients are often used to describe the relationship between variables; however, correlation does not imply independence. For finite discrete random variables, we can use the Pearson chi-square test to find independency. For the mixed type of continuous and discrete random variables, we do not have a general type of independent test. In this study, we develop a independence test of a continuous random variable and a discrete random variable without assuming a specific distribution using kernel density estimation. We provide some statistical criteria to test independence under some special settings and apply the proposed independence test to Pima Indian diabetes data. Through simulations, we calculate false positive rates and true positive rates to compare the proposed test and Kolmogorov-Smirnov test.

• Ocean Systems Engineering
• /
• 제7권3호
• /
• pp.275-298
• /
• 2017

This paper provides a practical stochastic method by which the maximum equilibrium scour depth below a pipeline exposed to random waves plus a current on mild slopes can be derived. The approach is based on assuming the waves to be a stationary narrow-band random process, adopting the Battjes and Groenendijk (2000) wave height distribution for mild slopes including the effect of breaking waves, and using the empirical formulas for the scour depth on the horizontal seabed by Sumer and Fredsøe (1996). The present approach is valid for wave-dominant flow conditions. Results for random waves alone and random wave plus currents have been presented and discussed by varying the seabed slope and water depth. An approximate method is also proposed, and comparisons are made with the present stochastic method. For random waves alone it appears that the approximate method can replace the stochastic method, whereas the stochastic method is required for random waves plus currents. Tentative approaches to related random wave-induced scour cases for random waves alone are also suggested.

• 산업경영시스템학회지
• /
• 제41권2호
• /
• pp.117-132
• /
• 2018

There always exist nonzero inspection errors whether inspectors are humans or automatic inspection machines. Inspection errors can be categorized by two types, type I error and type II error, and they can be regarded as either a constant or a random variable. Under the assumption that two types of random inspection errors are distributed with the "uniform" distribution on a half-open interval starting from zero, it was proved that inspectors overestimate any given fraction defective with the probability more than 50%, if and only if the given fraction defective is smaller than a critical value, which depends upon only the ratio of a type II error over a type I error. In addition, it was also proved that the probability of overestimation approaches one hundred percent as a given fraction defective approaches zero. If these critical phenomena hold true for any error distribution, then it might have great economic impact on commercial inspection plans due to the unfair overestimation and the recent trend of decreasing fraction defectives in industry. In this paper, we deal with the same overestimation problem, but assume a "symmetrical triangular" distribution expecting better results since our triangular distribution is closer to a normal distribution than the uniform distribution. It turns out that the overestimation phenomenon still holds true even for the triangular error distribution.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2003년도 추계 학술발표회 논문집
• /
• pp.73-78
• /
• 2003

This presentation derives a distribution function of the terminal value and running maximum of two-dimensional Brownian motion {X(t) = (X$_1$(t), X$_2$(T))', t > 0}. One random variable of the joint distribution is the terminal time value of the Brownian motion {X$_1$(t), t > 0}. The other random variable is the partial-time running maximum of the Brownian motion {X$_2$(t), t > 0}. With this distribution function, this presentation also derives an explicit pricing formula for a barrier option whose monitoring period of the option starts at an arbitrary date and ends at another arbitrary date before maturity.

• 산업경영시스템학회지
• /
• 제13권21호
• /
• pp.133-141
• /
• 1990

Human blood is a perishable product : it has a legal lifetime of 21 days from collection, during which it can be used for transfusion to a Patient of the same type, and after which it has to be discarded. Therefore, blood must be supplied safely and effectively because it is one of the medical resources which keep humanlife. In this study, the effects of blood issuing policies on average inventory levels and average age of blood at transfusion are determined by simulation applied the theory of absorbing Markov chains. And as a practical study, the daily demand distribution of blood is estimated by using the data of B General Hospital. The distribution estimated follows poisson distribution and the estimator of parameter estimated from the poisson distribution is 0.762. Simulation is done by using the parameter. The most important problem when control blood is the amount of outdata. So we compared random policy with Modified LIFO and Modified FIFO by using outdata. As a results it is shown that Modified LIFO and Modified FIFO by using outdata. As a results it Is shown that Modified LIFO and Modified FIFO present better issuing policy than Random Policy.

• 응용통계연구
• /
• 제25권6호
• /
• pp.1037-1047
• /
• 2012

The generalized linear mixed model(GLMM) is widely used in fitting categorical responses of clustered data. In the numerical approximation of likelihood function the normality is assumed for the random effects distribution; subsequently, the commercial statistical packages also routinely fit GLMM under this normality assumption. We may also encounter departures from the distributional assumption on the response variable. It would be interesting to investigate the impact on the estimates of parameters under misspecification of distributions; however, there has been limited researche on these topics. We study the sensitivity or robustness of the maximum likelihood estimators(MLEs) of GLMM for counts data when the true underlying distribution is normal, gamma, exponential, and a mixture of two normal distributions. We also consider the effects on the MLEs when we fit Poisson-normal GLMM whereas the outcomes are generated from the negative binomial distribution with overdispersion. Through a small scale Monte Carlo study we check the empirical coverage probabilities of parameters and biases of MLEs of GLMM.