Journal of Advanced Mathematical Modeling
Shahid Chamran University,Ahvaz





Vol. 1 , No. 2 

Spatial Logistic and Probit Regression Models for Analysis of
Frosting Data in Mazandaran Province
Abstract
Logistic and Probit regression models are usually used
in binary response variable analysis based on
independence assumption of the observations. But, in
practice, there are many situations in which, due to
their different locations in the underling space of
study, this assumption dose not satisfies. In spatial
statistics, it is generally supposed that the binary
observations are analyzed with indicator Kriging. In
this paper, we considered the spatial Logistic and
Probit regression models with auto correlated errors on
a rectangular grid. Also, in a simulation study, the
prediction accuracy of the models has been compared.
Finally, the implementation of the models for a
temperature data set, reported by weather stations in
Mazandaran province of Iran, is shown.
Keywords:
Spatial binary data, Logistic and Probit regression,
Rectangular grids. 


1 




Powers of Irreducible Characters of Finite Groups
Abstract
Let be an
irreducible character of a nonabelian group G.
For nonnegative integers n, m such that
, we study the
case when all the irreducible constituents ofare
linear. Mann proved that if G is a finite
nonabelian group with an irreducible character
such
that all the irreducible constituents of
are
linear, then
and as
a consequence G is nilpotent. In this paper we
generalize the result of Mann and prove that if m,n
are nonnegative integers with
, and if
is an
irreducible character of G, then all the
irreducible constituents of
are
linear if and only if
.
Keywords:
Character, Finite Groups, Irreducible Character, Power,
Product of Characters. 


19 




Application a Modified Imperialist Competitive Algorithm for
Solving the Traveling Salesman Problem
Abstract
This paper proposes a modified Imperialist Competitive
Algorithm (MICA) for solving the Traveling Salesman
Problem (TSP) that is different with common Imperialist
Competitive Algorithm (ICA) in assimilation policy
between Imperialist and colonies countries and
revolution of colonies. Furthermore, the 3opt local
search is used for increasing performance of the
algorithm. The new ICA algorithm is tested on nineteen
instances of TSBLIB and its performance is compared
withICA, Genetic Algorithm (GA), Particle Swarm
Optimization (PSO), Evolutionary Algorithm (EA) and Bee
Colony Optimization (BCO). Extensive computational tests
confirm the effectiveness of the proposed approach.
Keywords:
Traveling Salesman Problem, Imperialist Competitive
Algorithm, NPHard Problems. 


29 





Simulation of the breathing gases in the airways
Abstract
In this study, we have simulated the breathing of gases
in the human airways, and the simulated equation is
solved in two different ways. It is observed that the
numerical solution is more complicated in comparison
with the analytic one.
Keywords:
Analytic method; Numerical method; Concentration; Airway
shape; Trachea. 


51 





Goodnessoffit tests for the weighted exponential distribution
Abstract
In the new class of weighted
exponential distributions was presented by Gupta and
Kundu [1],
the skewness parameter has been added to the exponential
distribution. Therefore the
weighted exponential distribution has the
skewness and scale parameters. In this paper, we first
study Anderson and KolmogorovSmirnov
goodness of fit tests for this
class with unknown parameters.
Then, we apply bootstrap method for estimation of
Anderson’s quantile and another method for KolmogorovSmirnov.
We use the maximum likelihood method for
estimation of parameters. Finally, we
compare KolmogorovSmirnov and Anderson tests in a Monte
Carlo simulation study.The results show that the
Kolmogorov–Smirnov test has greater power than Anderson
test.
Keywords :
Goodness of fit test, Parametric bootstrap, Empirical
distribution function, Monte Carlo simulation. 


67 





Bayesian Analysis of RandomIntercept Models with the
SkewLaplace Distribution
Abstract
In fitting randomintercept models, it is commonly
assumed that the random effects and the error terms
follow the normal distribution. In many empirical
applications, the true distribution of data obeys
nonnormality and thus the main concern of most recent
studies is the use of alternative distributions. In this
paper, we propose a new class of randomintercept models
using the SkewLaplace distribution. The new regression
model is flexible in the analysis of correlated data and
simple in the implementation of Markov Chain Monte Carlo
methods, such as the Gibbs sampling approach. Using the
stochastic representation of the SkewLaplace
distribution we derive the full conditional posteriors
distributions in order to present the Bayesian inference
of model parameters. A real data analysis is illustrated
from the economic contexts to show the usefulness of the
proposed model.
Keywords:
Full conditional posterior density, Gibbs sampling,
Hierarchical representation, Latent variable, Random
effect.



85 



