• International Journal of Reliability and Applications
• /
• v.4 no.3
• /
• pp.97-111
• /
• 2003

By applying Theorem 2.6.4 (Fang and Zhang, 1990, p.66) the dispersion matrix of a multivariate power exponential (MPE) distribution is derived. It is shown that the MPE and the gamma distributions are related and thus the MPE and chi-square distributions are related. By extending Fang and Xu's Theorem (1987) from the normal distribution to the Univariate Power Exponential (UPE) distribution an explicit expression is derived for calculating the probability of an UPE random variable over an interval. A representation of the characteristic function (c.f.) for an UPE distribution is given. Based on the MPE distribution the probability density functions of the generalized non-central chi-square, the generalized non-central t, and the generalized non-central F distributions are derived.

• Journal of the Korean Statistical Society
• /
• v.34 no.4
• /
• pp.311-329
• /
• 2005

Skewed t distributions have attracted significant attention in the last few years. In this paper, a generalization - referred to as the skewed generalized t distribution - with the pdf f(x) = 2g(x)G(${\lambda}x$) is introduced, where g(${\cdot}$) and G (${\cdot}$) are taken, respectively, to be the pdf and the cdf of the generalized t distribution due to McDonald and Newey (1984, 1988). Several particular cases of this distribution are identified and various representations for its moments derived. An application is provided to rainfall data from Orlando, Florida.

• Communications for Statistical Applications and Methods
• /
• v.11 no.3
• /
• pp.435-446
• /
• 2004

In this paper we introduced a new score generating function for the rank dispersion function in a multiple linear model. Based on the new score function, we derived the asymptotic relative efficiency, ARE(11, rs), of our score function with respect to the Wilcoxon scores for the generalized F distributions which show very flexible distributions with a variety of shape and tail behaviors. We thoroughly explored the selection of r and s of our new score function that provides improvement over the Wilcoxon scores.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.5
• /
• pp.11-16
• /
• 1998

We analyze the software reliability growth models for the specified period from the viewpoint of theory of differential equations. we defien a genralized form of reliability growth models as follws: dN(t)/dt = b(t)f(N(t)), Where N(t) is the number of remaining faults and b(t) is the failure rate per software fault at time t. We show that the well-known three software reliability growth models - Goel - Okumoto, s-shaped, and Musa-Okumoto model- are special cases of the generalized form. We, also, extend the generalized form into an extended form being dN(t)/dt = b(t, .gamma.)f(N(t)), The genneralized form can be obtained if the distribution of failures is given. The extended form can be used to describe a software reliabilit growth model having weibull density function as a fault exposure rate. As an application of the generalized form, we classify three mentioned models according to the forms of b(t) and f(N(t)). Also, we present a case study applying the generalized form.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.5
• /
• pp.1829-1831
• /
• 2012

Background: The generalized gamma distribution statistics constitute an extensive family that contains nearly all of the most commonly used distributions including the exponential, Weibull and log normal. A saturated version of the model allows covariates having effects through all the parameters of survival time distribution. Accelerated failure-time models assume that only one parameter of the distribution depends on the covariates. Methods: We fitted both the conventional GG model and the saturated form for each of its members including the Weibull and lognormal distribution; and compared them using likelihood ratios. To compare the selected parameter distribution with log logistic distribution which is a famous distribution in survival analysis that is not included in generalized gamma family, we used the Akaike information criterion (AIC; r=l(b)-2p). All models were fitted using data for 369 women age 50 years or more, diagnosed with stage IV breast cancer in BC during 1990-1999 and followed to 2010. Results: In both conventional and saturated parametric models, the lognormal was the best candidate among the GG family members; also, the lognormal fitted better than log-logistic distribution. By the conventional GG model, the variables "surgery", "radiotherapy", "hormone therapy", "erposneg" and interaction between "hormone therapy" and "erposneg" are significant. In the AFT model, we estimated the relative time for these variables. By the saturated GG model, similar significant variables are selected. Estimating the relative times in different percentiles of extended model illustrate the pattern in which the relative survival time change during the time. Conclusions: The advantage of using the generalized gamma distribution is that it facilitates estimating a model with improved fit over the standard Weibull or lognormal distributions. Alternatively, the generalized F family of distributions might be considered, of which the generalized gamma distribution is a member and also includes the commonly used log-logistic distribution.

• Journal of the Chungcheong Mathematical Society
• /
• v.24 no.4
• /
• pp.795-805
• /
• 2011

In this paper, we study a family of continuous bivariate distributions that possesses the negative quadrant dependence property and the generalized negatively quadrant dependent F-G-M copula. We also develop the partial ordering of this new parametric family of negative quadrant dependent distributions.

• Journal of Korean Society for Quality Management
• /
• v.25 no.3
• /
• pp.11-21
• /
• 1997

The mean residual life(MRL) function gives the expected remaining life of a item at age t. In particular F is said to be an increasing intially then decreasing MRL(IDMRL) distribution if there exists a turing point $t^*\ge0$ such that m(s)$\le$ m(t) for 0$$\le s$\le$ t $t^*$, m(s)$\ge$ m(t) for $t^*\le$ s$\le$ t. If the preceding inequality is reversed, F is said to be a decreasing initially then increasing MRL(DIMRL) distribution. Hawkins, et al.(1992) proposed test of H0 : F is exponential versus$H_1$: F is IDMRL, and $H_0$ versus $H_1$' : F is DIMRL when turning point is unknown. Their test is based on a complete random sample $X_1$, …, $X_n$ from F. In this paper, we generalized Hawkins-Kochar-Loader test to random censored data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.212-217
• /
• 2012

A fuzzy set $A$ defined on a probability space ${\Omega}$, $\mathfrak{F}$, $P$ is called a fuzzy event. Zadeh defines the probability of the fuzzy event $A$ using the probability $P$. We define the generalized triangular fuzzy set and apply the extended algebraic operations to these fuzzy sets. A generalized triangular fuzzy set is symmetric and may not have value 1. For two generalized triangular fuzzy sets $A$ and $B$, $A(+)B$ and $A(-)B$ become generalized trapezoidal fuzzy sets, but $A({\cdot})B$ and $A(/)B$ need not to be a generalized triangular fuzzy set or a generalized trapezoidal fuzzy set. We define the normal fuzzy probability on $\mathbb{R}$ using the normal distribution. And we calculate the normal fuzzy probability for generalized triangular fuzzy sets.

• The Korean Journal of Applied Statistics
• /
• v.24 no.5
• /
• pp.931-939
• /
• 2011

In practice, thresholds are determined by the two subjective assessment methods in a generalized pareto distribution of mean extreme function(MEF-graph) or Hill-graph. To remedy the problem of subjectiveness of these methods, we propose an alternative method to determine the threshold based on the robust statistics. We compared the MEF-graph, Hill-graph and our method through VaRs on the Korean stock market data from January 5, 1987 to August 3, 2009. As a result, the VaR based on the proposed method is not much different from the existing methods, and the standard deviation of VaR for our method was the smallest. The results show that our method can be a promising alternative to determine thresholds of the generalized pareto distributions.

• Journal of Forest and Environmental Science
• /
• v.36 no.1
• /
• pp.7-16
• /
• 2020

Tree size distribution modelling is an integral part of forest management. Most distribution yield systems rely on some flexible probability models. In this study, a simple finite mixture of two components two-parameter Weibull distribution was compared with complex four-parameter distributions in terms of their fitness to predict tree size distribution of teak (Tectona grandis Linn f) plantations. Also, a system of equation was developed using Seemingly Unrelated Regression wherein the size distributions of the stand were predicted. Generalized beta, Johnson's SB, Logit-Logistic and generalized Weibull distributions were the four-parameter distributions considered. The Kolmogorov-Smirnov test and negative log-likelihood value were used to assess the distributions. The results show that the simple finite mixture outperformed the four-parameter distributions especially in stands that are bimodal and heavily skewed. Twelve models were developed in the system of equation-one for predicting mean diameter, seven for predicting percentiles and four for predicting the parameters of the finite mixture distribution. Predictions from the system of equation are reasonable and compare well with observed distributions of the stand. This simplified mixture would allow for wider application in distribution modelling and can also be integrated as component model in stand density management diagram.