|
Temporal Changes in the Parameters of Statistical Distribution
of Journal Impact Factor
-- S K Mishra
The statistical distribution of Journal Impact Factor (JIF) is characteristically asymmetric and
non-mesokurtic. Even the distribution of
log10(JIF) exhibits conspicuous skewness and
non-mesokurtocity. This paper estimates the parameters of Johnson SU distribution fitting to the
log10(JIF) data for 10 years, 1999-2008, and studies the temporal variations in those estimated parameters. The
`over-the-samples stability' in the estimated parameters for each year is also studied by the method of resampling close
to bootstrapping. It has been found that
log10(JIF) is Pearson Type-IV distributed. Johnson SU distribution fits very well to the data and yields parameters stable over the samples. The paper concludes
that Johnson SU distribution is the best choice to fit to the
log10(JIF) data. It is also found that over the
years the log10(JIF) distribution has become more skewed and leptokurtic, possibly suggesting the
Matthew effect in operation, which means that more cited journals are cited even more over time.
© 2010 IUP. All Rights Reserved.
An EOQ Model with Pareto Distribution for Deterioration, Trapezoidal Type Demand and Backlogging Under Trade Credit Policy
-- Narayan Singh, Bindu Vaish and S R Singh
This paper studies and analyzes an inventory model with the assumption that the lifetime of the
commodity is random and follows a generalized Pareto distribution. It is also assumed that demand rate is
of Trapezoidal type, i.e., the demand rate is a piecewise linear function and this demand rate is used
when stock is available as well as during shortage period. Shortage is completely backlogged and
backlogging rate is of Trapezoidal type. This study considers the traditional hypothesis that at the end of
credit period, the retailer makes a partial payment of the total purchasing cost and pays off the
remaining balance by loan from bank. With suitable cost consideration, the total cost function is obtained.
Minimizing total cost function, the optimal ordering quantity and optimal time are obtained. The numerical
examples are presented to illustrate the model.
© 2010 IUP. All Rights Reserved.
A Survey of Ridge Regression for Improvement
Over Ordinary Least Squares
-- Rajeshwar Singh
Multicollinearity may be a possible cause in case of study with two or more explanatory variables. In
the presence of multicollinearity, the design matrix becomes nearly singular and hence X and the corresponding are not of full rank. In this situation, the Ordinary Least Square (OLS) estimate cannot be
obtained. Thus, adequate attention is required to be given on the presence of multicollinearity in the data. In
this survey only ridge regression is discussed as a solution to the problem of multicollinearity. Hoerl
and Kennard proposed the technique of ridge regression that has become a popular tool for data
analysis faced with the problem of a high degree of multicollinearity. They have suggested adding a small
positive quantity in the diagonal elements of the design matrix,
before inverting it. In other words, they proposed
in place of where and are the ridge and OLS estimates of the parameter vector b respectively. Though the ridge estimate is biased, it has a
smaller mean squared error than OLS estimator. A critical appraisal is also given on the choice of biasing
parameter, in addition to its properties, its relation with other estimators and Bayesian and
non-Bayesian interpretations.
© 2010 IUP. All Rights Reserved.
|
December '10
