1
1735-1294
Statistical Research and Training Center - Statistical Centre of Iran
168
General
Dynamic Bayesian Information Measures
Ebrahimi
Nader
Kirmani
S. N. U. A.
Soofi
S.
1
3
2007
3
2
113
138
13
02
2016
13
02
2016
This paper introduces measures of information for Bayesian analysis when the support of data distribution is truncated progressively. The focus is on the lifetime distributions where the support is truncated at the current age t>=0. Notions of uncertainty and information are presented and operationalized by Shannon entropy, Kullback-Leibler information, and mutual information. Dynamic updatings of prior distribution of the parameter of lifetime distribution based on observing a survival at age t and observing a failure or the residual lifetime beyond t are presented. Dynamic measures of information provided by the data about the parameter of lifetime distribution, and dynamic predictive information are introduced. These measures are applied to two well-known lifetime models. The paper concludes with some remarks on use of generalized uncertainty and information measures, and some topics for further research.
164
General
Comparison of Estimates Using Record Statistics from Lomax Model: Bayesian and Non Bayesian Approaches
Ellah
H.
1
3
2007
3
2
139
158
13
02
2016
13
02
2016
This paper address the problem of Bayesian estimation of the parameters, reliability and hazard function in the context of record statistics values from the two-parameter Lomax distribution. The ML and the Bayes estimates based on records are derived for the two unknown parameters and the survival time parameters, reliability and hazard functions. The Bayes estimates are obtained based on conjugate prior for the scale parameter and discrete prior for the shape parameter of this model. This is done with respect to both symmetric loss function (squared error loss), and asymmetric loss function (linear-exponential (LINEX)) loss function. The maximum likelihood and the different Bayes estimates are compared via Monte Carlo simulation study. A practical example consisting of real record values including in the data from an accelerated test on insulating fluid reported by Nelson was used for illustration and comparison. Finally, Bayesian predictive density function, which is necessary to obtain bounds for predictive interval of future record is derived and discussed using a numerical example.
166
General
Bivariate Semi-Logistic Distribution and Processes
Mathew
Thomas
Jayakumar
K.
1
3
2007
3
2
159
176
13
02
2016
13
02
2016
Bivariate semi-logistic and Marshall-Olkin bivariate semi-logistic distributions are introduced. Some properties of these distributions are studied. First order autoregressive processes with bivariate semi-logistic and Marshall-Olkin bivariate semi-logistic distributions as marginals are introduced and studied.
167
General
On Symmetric Extended Generalized Logistic Distribution
Olapade
K.
1
3
2007
3
2
190
177
13
02
2016
13
02
2016
In this paper, we consider a form of the generalized logistic distribution named symmetric extended generalized logistic distribution or extended type III generalized logistic distribution. The distribution is derived by compounding a two-parameter generalized Gumbel distribution with a two-parameter generalized gamma distribution. The cumulative distribution and some properties of this distribution like moments and related statistics are established. Some theorems that characterize the distribution are stated and proved. Estimation of the parameters and an application of the distribution are also presented.
165
General
Symmetrised Doubly Non Central, Nonsingular Matrix Variate Beta Distribution
García
Díaz-
Jáimez
Gutiérrez-
1
3
2007
3
2
191
202
13
02
2016
13
02
2016
In this paper, we determine the symmetrised density of a nonsingular doubly noncentral matrix variate beta type I and II distributions under different definitions.
169
General
Numerical Methods of Option Pricing for Two Specific Models of Electricity Prices
Zamani
Shiva
1
3
2007
3
2
203
221
13
02
2016
13
02
2016
In this work, two models are proposed for electricity prices as energy commodity prices which in addition to mean-reverting properties have jumps and spikes, due to non-storability of electricity. The models are simulated using an Euler scheme, and then the Monte-Carlo method is used to estimate the expectation of the discounted cash-flow under historical probability, which is considered as the option price. A so called random variable simulation and a control variate method are then used to decrease, the discretization error and the Monte-Carlo error, respectively. As the option prices satisfy PDE's associated with the models, by solving these PDE's, numerically, we can find the option prices by a second method, thereby being able to make comparisons.