• Communications for Statistical Applications and Methods
• /
• v.7 no.1
• /
• pp.141-150
• /
• 2000

In this paper we propose a method of computing Bayes factor to detect an outlier in a random effects model. When no information is available and hence improper noninformative priors should be used Bayes factor includes the unspecified constants and has complicated computational burden. To solve this problem we use the fractional Bayes factor (FBF) of O-Hagan(1995) and the generalized Savage0-Dickey density ratio of Verdinelli and Wasserman (1995) The proposed method is applied to outlier deterction problem We perform a simulation of the proposed approach with a simulated data set including an outlier and also analyze a real data set.

• Proceedings of the Safety Management and Science Conference
• /
• 2011.04a
• /
• pp.695-699
• /
• 2011

The study presents an efficient design method of Venn Diagram that can be used when implementing the quality design of experiments based on generalizability theory. The paper examines four mixed and combined models that are designed by fixed factor, random factor, crossed factor and nested factor. The models considered in this research are $A^*{\times}B^*{\times}C$, (B: $A^*$)${\times}C$, $A{\times}B{\times}C$ and (B: A)${\times}C$.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.464-467
• /
• 2001

A good arithmetic random number generator should possess full period, uniformity and independence, etc. To obtain the excellent random number generator, many researchers have found good parameters. Also an initial seed is the important factor in random number generator. But, there is no theoretical guideline for using the initial seeds. Therefore, random number generator is usually used with the arbitrary initial seed. Through the empirical tests, we show that the choice of the initial values for the seed is important to generate good random numbers.

• Proceedings of the KSME Conference
• /
• 2000.11a
• /
• pp.141-146
• /
• 2000

Crack closure and growth behavior of physically short fatigue cracks under random loading are investigated by performing narrow- and wide-band random loading tests for various stress ratios. The importance of the crack closure phenomenon is examined by predicting the growth lives of short cracks using obtained crack opening behavior. Artificially prepared two-dimensional, short through-thickness cracks are used. The crack opening load of short cracks is much lower under random loading than under constant-amplitude loading corresponding to the largest load cycle in a random load history. This result indicates that the largest load cycle in a random load history has an effect to enhance crack opening of short cracks. Most of the life prediction ratios are within the factor of 2 scatter band except several data at very short crack sizes, indicating that crack growth predictions based on the measured crack opening data are excellent. From the results obtained in this study, it can be concluded that crack closure is the primary factor governing fatigue crack growth of short cracks under random loading as well as under constant-amplitude loading.

• Proceedings of the KSME Conference
• /
• 2004.11a
• /
• pp.208-213
• /
• 2004

Methods for estimation of the effective stress intensity factor range (${\Delta}K_{eff}$) are evaluated for narrow and wide band random loading crack growth test data of 2024-T351 aluminum alloy. Three methods of determining $K_{op}$, visual measurement, ASTM offset compliance method, and the neural network method proposed by Kang and Song, and three methods of estimating ${\Delta}K_{eff}$, conventional, the 2/PI0 and 2/PI methods proposed by Donald and Paris, are compared in a quantitative manner by using the results of fatigue crack growth life prediction under random loading. For all $K_{op}$ determination methods discussed, the 2/PI0 and 2/PI methods of estimating ${\Delta}K_{eff}$ provide better results than conventional method for narrow and wide band random loading data.

• Geomechanics and Engineering
• /
• v.33 no.6
• /
• pp.543-552
• /
• 2023

The design of foundations based on a deterministic approach may not be safe and reliable occasionally, since soils sometimes show considerable spatial variability, and thus, significant uncertainties in turn affect the estimation of footing bearing capacity. The design of footing on cohesionless stratums on the basis of reliability analysis has not received much attention. This paper performs two-dimensional random finite difference analyses of shallow strip footings on a spatially variable frictional soil considering correlation structure. Friction angle (ϕ) is considered as a log-normally distributed random variable and Monte Carlo Simulation is then performed to determine the statistical response based on the random fields. A new approach reliability-based safety factor is defined based on various reliability levels by considering the coefficient of variation of ϕ and correlation length in both the horizontal and vertical directions. The comparison of the probabilistic safety factor and the conventional one illustrates the limitations of the deterministic safety factor and provides insight into how the heterogeneity of soils properties affects the required safety factor. Results show that the conventional safety factor of 3 can be conservative in some cases, especially for soil with low values of mean ϕ and COV_ϕ.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.2
• /
• pp.191-204
• /
• 2010

There is a well-defined propagation model, named the random constant spread (RCS) model, which explains worms that spread their clones with a random scanning strategy. This model uses the number of infected hosts in a domain as a factor in the worms' propagation. However, there are difficulties in explaining the characteristics of new Internet worms because they have several considerable new features: the denial of service by network saturation, the utilization of a faster scanning strategy, a smaller size in the worm's propagation packet, and to cause maximum damage before human-mediated responses are possible. Therefore, more effective factors are required instead of the number of infected hosts. In this paper, the network bandwidth usage rate is found to be an effective factor that explains the propagations of the new Internet worms with the random scanning strategy. The analysis and simulation results are presented using this factor. The simulation results show that the scan rate is more sensitive than the propagation packet for detecting worms' propagations.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.4
• /
• pp.727-733
• /
• 2011

To understand the stock market structure it is very important to extract meaningful information by analyzing the correlation matrix between stock returns. Recently there has been many studies on the correlation matrix using the Random Matrix Theory. In this paper we adopt this random matrix methodology to a single-factor model and we obtain meaningful information on the correlation matrix. In particular we observe the analysis of the correlation matrix using the single-factor model explains the real market data and as a result we confirm the usefulness of the single-factor model.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.339-345
• /
• 2004

This paper deals with a mixed logit model for ordered polytomous data. There are two types of factors affecting the response varable in this paper. One is a fixed factor with finite quantitative levels and the other is a random factor coming from an experimental structure such as a randomized complete block design. It is discussed how to set up the model for analyzing ordered polytomous data and illustrated how to estimate the paramers in the given model.

• 제어로봇시스템학회:학술대회논문집
• /
• 1993.10b
• /
• pp.451-455
• /
• 1993

Interaction between system and disturbance results in system with time-dependent parameter. Parameter variation due to interaction has random characteristics. Most of the randomly varying parameters in control problem is regarded as white noise random process which is not a realistic model. In real situation those random variation is colored noise random process. Modified F-P-K equation is proposed to get the response of the random parametric system using some correction factor. Proposed technique is employed to obtain the colored noise parametric system response and confirmed via Monte-Carlo Simulation.