-
On Nonnegative Integer-Valued Lévy Processes and Applications in Probabilistic Number Theory and Inventory Policies
Huiming Zhang,
Jiao He,
Hanlin Huang
Issue:
Volume 2, Issue 5, September 2013
Pages:
110-121
Received:
3 August 2013
Published:
30 August 2013
Abstract: Discrete compound Poisson processes (namely nonnegative integer-valued Lévy processes) have the property that more than one event occurs in a small enough time interval. These stochastic processes produce the discrete compound Poisson distributions. In this article, we introduce ten approaches to prove the probability mass function of discrete compound Poisson distributions, and we obtain seven approaches to prove the probability mass function of Poisson distributions. Finally, we discuss the connection between additive functions in probabilistic number theory and discrete compound Poisson distributions and give a numerical example. Stuttering Poisson distributions (a special case of discrete compound Poisson distributions) are applied to numerical solution of optimal (s, S) inventory policies by using continuous approximation method.
Abstract: Discrete compound Poisson processes (namely nonnegative integer-valued Lévy processes) have the property that more than one event occurs in a small enough time interval. These stochastic processes produce the discrete compound Poisson distributions. In this article, we introduce ten approaches to prove the probability mass function of discrete comp...
Show More
-
Efficiency of Neyman Allocation Procedure over other Allocation Procedures in Stratified Random Sampling
Olayiwola Olaniyi Mathew,
Apantaku Fadeke Sola,
Bisira Hammed Oladiran,
Adewara Adedayo Amos
Issue:
Volume 2, Issue 5, September 2013
Pages:
122-127
Received:
6 August 2013
Published:
30 August 2013
Abstract: In sampling, we have interest in precision and in order to create the precision, we make use of prior knowledge of the population. We try to put the population into series of homogeneous groups and by this, the precision will be increased. When the population of interest can be divided into k homogeneous groups and the sample of observation is taken from each group, we have a stratified random sample and each group is called a stratum. The study was therefore designed to investigate the efficiency of Neyman allocation procedure over equal and proportional allocations. The data used for this research were primary data collected from ten Markets in Abeokuta, Ogun State, Nigeria on the prices of Peak Milk (Nigeria made). A stratified random sampling scheme was used in selecting 10 markets in Abeokuta, Ogun State, Nigeria. Each market stands as a stratum. From each stratum, independent sample was selected randomly based on equal, proportional and Neyman/Optimum allocation procedures. Statistic was obtained from each stratum and combined estimate of the separate statistic was also obtained for each of the allocation procedure. Considering the analysis and estimates obtained, the mean and variance under Neyman allocation procedure were 1356.672 and 21.45 respectively. For proportional allocation, the mean was 1349.3069 and variance was 38.98 while equal allocation gave mean of 1352 and variance of 170.3238. Neyman/Optimum allocation procedure gave the least variance. This was followed by Proportional allocation and Equal allocation. Neyman allocation procedure is the best selection procedure. Hence, for estimating the average and the variance of the prices of Peak Milk (Nigeria Made) in the markets in Abeokuta, of all the three sample allocation procedures considered in this paper, Neyman allocation procedure is the best and hence the most efficient.
Abstract: In sampling, we have interest in precision and in order to create the precision, we make use of prior knowledge of the population. We try to put the population into series of homogeneous groups and by this, the precision will be increased. When the population of interest can be divided into k homogeneous groups and the sample of observation is take...
Show More
-
Bayesian Estimation Using MCMC Approach Based on Progressive First-Failure Censoring from Generalized Pareto Distribution
Mohamed Abdul Wahab Mahmoud,
Ahmed Abo-Elmagd Soliman,
Ahmed Hamed Abd Ellah,
Rashad Mohamed El-Sagheer
Issue:
Volume 2, Issue 5, September 2013
Pages:
128-141
Received:
17 August 2013
Published:
30 August 2013
Abstract: In this paper, based on a new type of censoring scheme called a progressive first-failure censored, the maximum likelihood (ML) and the Bayes estimators for the two unknown parameters of the Generalized Pareto (GP) distribution are derived. This type of censoring contains as special cases various types of censoring schemes used in the literature. A Bayesian approach using Markov Chain Monte Carlo (MCMC) method to generate from the posterior distributions and in turn computing the Bayes estimators are developed. Point estimation and confidence intervals based on maximum likelihood and bootstrap methods are also proposed. The approximate Bayes estimators have been obtained under the assumptions of informative and non-informative priors. A numerical example is provided to illustrate the proposed methods. Finally, the maximum likelihood and different Bayes estimators are compared via a Monte Carlo simulation study.
Abstract: In this paper, based on a new type of censoring scheme called a progressive first-failure censored, the maximum likelihood (ML) and the Bayes estimators for the two unknown parameters of the Generalized Pareto (GP) distribution are derived. This type of censoring contains as special cases various types of censoring schemes used in the literature. A...
Show More
-
Optimum Allocation of Multi-Items in Stratified Random Sampling Using Principal Component Analysis
Apantaku Fadeke Sola.,
Olayiwola Olaniyi Mathew,
Adewara Amos Adedayo
Issue:
Volume 2, Issue 5, September 2013
Pages:
142-148
Received:
6 August 2013
Published:
10 September 2013
Abstract: The problem of allocation with more than one characteristic in stratified sampling is conflicting in nature, as the best allocation for one characteristic will not in general be best for others. Some compromise must be reached to obtain an allocation that is efficient for all characteristics. In this study, the allocation of a sample to strata which minimizes cost of investigation, subject to a given condition about the sampling error was considered. The data on four socioeconomic characteristics of 400 heads of households in Abeokuta South and Ijebu North Local Government Areas (LGAs) of Ogun State, Nigeria were investigated. These comprised of 200 households from each LGA. The characteristics were occupation, income, household size and educational level. Optimal allocation in multi-item was developed as a multivariate optimization problem by finding the principal components. This was done by determining the overall linear combinations that concentrates the variability into few variables. From the principal component analysis, it was seen that for both Abeokuta and Ijebu data sets, the variance based on the four characteristics as multivariate is less than that of the variables when considered as a univariate. From the results, it was seen that there was no difference in the percentage of the total variance accounted for by the different components from the merged sample when compared with the individual sample. Optimum allocation was achieved when there was stratification
Abstract: The problem of allocation with more than one characteristic in stratified sampling is conflicting in nature, as the best allocation for one characteristic will not in general be best for others. Some compromise must be reached to obtain an allocation that is efficient for all characteristics. In this study, the allocation of a sample to strata whic...
Show More
