• Journal of the Korean Data and Information Science Society
• /
• v.9 no.1
• /
• pp.47-56
• /
• 1998

Zero-Inflated Poisson models are mixed models of the Poisson and Bernoulli models. Recently Zero-Inflated Poisson distributions have been used frequently rather than previous Poisson distributions because the developement of industrial technology make few defects in manufacturing process. It is important that univariate Zero-Inflated Poisson distributions are extended to bivariate distributions to generalize the multivariate distributions. In this paper we proposed three types of the bivariate Zero-Inflated Poisson distributions and obtained these moments. We compared the three types of distributions by using the moments.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1231-1239
• /
• 2014

In this paper we analyse the distributions of the number of goals scored by home teams and away teams in K-league soccer outcomes between 1983 and 2012. Real soccer data is explained in K-league using statistical distributions such that Poisson, negative binomial, extreme value and zero inflated Poisson. How close the goals of home and away fits the different distributions are tested by performing chi-square goodness of fit tests. According to these tests, the Poisson distribution gives the best fit to the home goals data. But it is best to model the away goals data on zero inflated Poisson distribution. Also, there is some weak evidence of the dependence for home and away goals.

• Journal of applied mathematics & informatics
• /
• v.31 no.1_2
• /
• pp.111-118
• /
• 2013

External excitations are employed to investigate properties of optical media, with measurement data often analyzed via linear response theory. In this respect, external forcing is modeled here by well-known Poisson and negative-binomial distributions. Ensuing dynamics is examined with a special attention to the relative decay rates of damped harmonic oscillators to such external forcing, along with its relationship to other physical phenomena.

• Journal of the Korean Statistical Society
• /
• v.32 no.1
• /
• pp.21-24
• /
• 2003

In the recent papers by Harris and Park (1994) and by Hui and Park (2000), a family of lattice distributions derived from a sum of independent identically distributed random variables is examined. In this paper we generalize a result of Hui and Park (2000) on lattice distributions on the torus using the Poisson summation formula.

• Journal of the Korean Statistical Society
• /
• v.29 no.3
• /
• pp.315-318
• /
• 2000

In this note we use the Poisson Summation Formula to generalize a result of Harris and Park (1994) on lattice distributions induced by uniform (0,1) random variables to those generated by random variables with step functions as their probability functions.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.2
• /
• pp.619-631
• /
• 2006

When X and Y have independent Poisson distributions, we develop a Bayesian one-sided testing procedures for the ratio of two Poisson means. We propose the objective Bayesian one-sided testing procedures for the ratio of two Poisson means based on the fractional Bayes factor and the intrinsic Bayes factor. Some real examples are provided.

• 한국데이터정보과학회:학술대회논문집
• /
• 2006.04a
• /
• pp.295-306
• /
• 2006

When X and Y have independent Poisson distributions, we develop a Bayesian one-sided testing procedures for the ratio of two Poisson means. We propose the objective Bayesian one-sided testing procedures for the ratio of two Poisson means based on the fractional Bayes factor and the intrinsic Bayes factor. Some real examples are provided.

• IE interfaces
• /
• v.17 no.1
• /
• pp.1-12
• /
• 2004

In a modern semiconductor device manufacturing industry, statistical bin limits on wafer level test bin data are used for minimizing value added to defective product as well as protecting end customers from potential quality and reliability excursion. Most wafer level test bin data show skewed distributions. By Monte Carlo simulation, this paper evaluates methods and sample size effect regarding determination of statistical bin limits. In the simulation, it is assumed that wafer level test bin data follow the Poisson distribution. Hence, typical shapes of the data distribution can be specified in terms of the distribution's parameter. This study examines three different methods; 1) percentile based methodology; 2) data transformation; and 3) Poisson model fitting. The mean square error is adopted as a performance measure for each simulation scenario. Then, a case study is presented. Results show that the percentile and transformation based methods give more stable statistical bin limits associated with the real dataset. However, with highly skewed distributions, the transformation based method should be used with caution in determining statistical bin limits. When the data are well fitted to a certain probability distribution, the model fitting approach can be used in the determination. As for the sample size effect, the mean square error seems to reduce exponentially according to the sample size.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.6 no.2
• /
• pp.51-55
• /
• 1981

In both continuous-review and periodic-review non-stationary inventory systems, the non-stationary Poisson demand process and the associated inventory position processes were proved being mutually independent of each other, which lead to the probability distribution of the corresponding net inventory position process in the form of a finite product sum of those two process distributions. It is also discussed how these results can correspond to analytical stochastic inventory cost function formulations in terms of the probability distributions of the processes.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.4
• /
• pp.1209-1217
• /
• 2006

For the mixture of shifted Poisson distributions, a method of parameter estimation is proposed. The range of the shifted parameters are estimated first and for each shifted parameter set EM algorithm is applied to estimate the other parameters of the distribution. Among the estimated parameter sets, one with minimum likelihood for given data is to be set as the final estimate. In simulation experiments, the suggested estimation method shows to have a good performance.