• Journal of the Korea Society of Computer and Information
• /
• v.11 no.1 s.39
• /
• pp.45-53
• /
• 2006

Finite failure NHPP models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, Goel-Okumoto and Yamada-Ohba-Osaki model was reviewed, proposes the $x^2$ reliability model, which can capture the increasing nature of the failure occurrence rate per fault. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on SSE, AIC statistics and Kolmogorov distance, for the sake of efficient model, was employed. Analysis of failure using real data set, SYS2(Allen P.Nikora and Michael R.Lyu), for the sake of proposing shape parameter of the $x^2$ distribution using the degree of freedom, was employed. This analysis of failure data compared with the $x^2$ model and the existing model using arithmetic and Laplace trend tests, Kolmogorov test is presented.

• Journal of the Korea Computer Industry Society
• /
• v.6 no.5
• /
• pp.689-696
• /
• 2005

Finite failure NHPP models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, Goel-Okumoto and Yamada-Ohba-Osaki model was reviewed, proposes the Kappa(2) reliability model, which can capture the nomotonic decreasing nature of the failure occurrence rate per fault. Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on sum of the squared errors and Kolmogorov distance, for the sake of efficient model, was employed. Analysis of failure using real data set, SYS2(Allen P.Nikora and Michael R.Lyu), for the sake of proposing two parameter of the Kappa distribution, was employed. This analysis of failure data compared with the Kappa model and the existing model using arithmetic and Laplace trend tests, bias tests is presented.

• The Korean Journal of Applied Statistics
• /
• v.19 no.1
• /
• pp.1-12
• /
• 2006

The aim of this study is to analyse a survey data on the number of charitable donations using a mixture of two Poisson regression models. The survey was conducted in 2002 by Volunteer 21, an nonprofit organization, based on Koreans, who were older than 20. The mixture of two Poisson distributions is used to model the number of donations based on the empirical distribution of the data. The mixture of two Poisson distributions implies the whole population is subdivided into two groups, one with lesser number of donations and the other with larger number of donations. We fit the mixture of Poisson regression models on the number of donations to identify significant covariates. The expectation-maximization algorithm is employed to estimate the parameters. We computed 95% bootstrap confidence interval based on bias-corrected and accelerated method and used then for selecting significant explanatory variables. As a result, the income variable with four categories and the volunteering variable (1: experience of volunteering, 0: otherwise) turned out to be significant with the positive regression coefficients both in the lesser and the larger donation groups. However, the regression coefficients in the lesser donation group were larger than those in larger donation group.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.2
• /
• pp.395-400
• /
• 1997

An important quality characteristic of a software reliability.Software reliablilty growh models prvied the tools to evluate and moniter the reliabolty growth behavior of the sofwate during the testing phase Therefore failure data collected during the testing phase should be continmuosly analyzed on the basis of some selected software reliability growth models.For the cases where nonhomogeneous Poisson proxess models are the candiate models,we suggest Poisson regression model, which expresses the relationship between the expeted and actual failures counts in disjonint time intervals,for analyzing the failure count data.The weighted lest squares method is then used to-estimate the paramethers in the parameters in the model:The resulting estimators are equivalent to the maximum likelihood estimators. The method is illustrated by analyzing the failutr count data gathered from a large- scale switchong system.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.5 s.43
• /
• pp.9-18
• /
• 2006

Finite failure NHPP models presented in the literature exhibit either constant, monotonic increasing or monotonic decreasing failure occurrence rates per fault. In this paper, Goel-Okumoto and Yamada-Ohba-Osaki model was reviewed, proposes the exponentiated exponential distribution reliability model, which maked out efficiency substituted for gamma and Weibull model(2 parameter shape illustrated by Gupta and Kundu(2001) Algorithm to estimate the parameters used to maximum likelihood estimator and bisection method, model selection based on SSE, AIC statistics and Kolmogorov distance, for the sake of efficient model, was employed. Analysis of failure using NTDS data set for the sake of proposing shape parameter of the exponentiated exponential distribution was employed. This analysis of failure data compared with the exponentiated exponential distribution model and the existing model (using arithmetic and Laplace trend tests, bias tests) is presented.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.2
• /
• pp.387-397
• /
• 2015

In this paper, the determinants of the number of job changes in the SMEs (small and medium enterprises) youth-intern project is analysed, utilizing SMEs youth-intern DB and employment insurance DB. Since the number of job changes are count data which take integer values other than negative values, general linear regression analysis becomes inappropriate. Therefore, four models such as Poisson regression model, zero inflated Poisson regression model, negative binomial regression model and zero inflated negative binomial regression model are tried to fit count data. A zero inflated negative binomial regression model is selected to be the best model. Major results are the followings. First, the number of job changes is shown to be significantly smaller in the treatment group than in the control group. Second, the number of job changes turns out to be significantly smaller in the young-age group than in the old-age group. Third, it is also shown that the number of job changes of man is significantly greater than that of woman. Lastly, the number of job changes in the bigger firm is shown to be significantly less than that of the smaller firm.

• The Korean Journal of Applied Statistics
• /
• v.29 no.2
• /
• pp.345-354
• /
• 2016

Inhomogeneous Poisson process models are widely applied to landslide data to understand how environmental variables systematically influence the risk of landslides. However, those models cannot successfully explain the clustering phenomenon of landslide locations. In order to overcome this limitation, we propose to use a Cauchy cluster process model and show how it improves the goodness of fit to the landslide data in terms of K-function. In addition, a numerical study is performed to select the optimal estimation method for the Cauchy cluster process.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.2D
• /
• pp.181-189
• /
• 2008

The purpose of this study is to comparatively analyze the characteristics and models of traffic accidents by day and nighttime. In pursuing the above, this study gives particular attentions to testing the differences and developing the models (multiple linear and non-linear and Poisson and negative binomial regression) using the data of Cheongju 4-legged signalized intersections. The main results analyzed are as follows. First, the differences between day and nighttime accidents were defined. Second, 12 accident models which are all statistically significant were developed. Finally, the differences between day and nighttime models were comparatively analyzed using the common and specific variables.

• The KIPS Transactions:PartD
• /
• v.8D no.4
• /
• pp.393-400
• /
• 2001

This paper presents a stochastic model for the software failure phenomenon based on a nonhomogeneous Poisson process (NHPP). The failure process is analyzed to develop a suitable mean value function for the NHPP ; expressions are given for several performance measure. Actual software failure data are compared with generalized model by Goel dependent on the constant reflecting the quality of testing. The performance measures and parametric inferences of the new models, Rayleigh and Gumbel distributions, are discussed. The results of the new models are applied to real software failure data and compared with Goel-Okumoto and Yamada, Ohba and Osaki models. Tools of parameter inference was used method of the maximun likelihood estimate and the bisection algorithm for the computing nonlinear root. In this paper, using the sum of the squared errors, model selection was employed. The numerical example by NTDS data was illustrated.

• Journal of Navigation and Port Research
• /
• v.48 no.1
• /
• pp.62-77
• /
• 2024

In the maritime shipping industry, imbalance between supply and demand has persistently increased, leading to the utilization of blank sailings by major shipping companies worldwide as a key means of flexibly adjusting vessel capacity in response to shipping market conditions. Traditionally, blank sailings have been frequently implemented around the Chinese New Year period. However, due to unique circumstances such as the global pandemic starting in 2020 and trade tensions between the United States and China, shipping companies have recently conducted larger-scale blank sailings compared to the past. As blank sailings directly impact freight transport delays, they can have negative repercussions from perspectives of both businesses and consumers. Therefore, this study employed Poisson regression models and negative binomial regression models to analyze the influence of maritime freight rate determinants on shipping companies' decisions regarding blank sailings, aiming to proactively address potential consequences. Results of the analysis indicated that, in Poisson regression analysis for 2M, significant variables included global container shipping volume, container vessel capacity, container ship scrapping volume, container ship newbuilding index, and OECD inflation. In negative binomial regression analysis, ocean alliance showed significance with global container shipping volume and container ship order volume, the alliance with container ship capacity and interest rates, non-alliance with international oil prices, global supply chain pressure index, container ship capacity, OECD inflation, and total alliance with container ship capacity and interest rates.