Moments of order statistics of Topp–Leone distribution

We derive explicit algebraic expressions for both of the single and product moments of order statistics from Topp–Leone distribution. We also give an identity about single moments of order statistics. These expressions will be useful for computational purposes.     

How to improve the fit of Archimedean copulas by means of transforms

The selection of copulas is an important aspect of dependence modeling issues. In many practical applications, only a limited number of copulas is tested and the copula with the best result for a goodness-of-fit test is chosen, which, however, does not always lead to the best possible fit. In this paper we develop a practical and logical method for improving the goodness-of-fit of a particular Archimedean copula by means of transforms. In order to do this, we introduce concordance invariant transforms which can also be tail dependence preserving, based on an analysis on the λ-function, l = \fracjj￠{\lambda=\frac{\varphi}{\varphi'}}, where j{\varphi} is the Archimedean generator. The methodology is applied to the data set studied in Cook and Johnson (J R Stat Soc B 43:210–218, 1981) and Genest and Rivest (J Am Stat Assoc 88:1043–1043, 1993), where we improve the fit of the Frank copula and obtain statistically significant results.     

SB-robust estimator for the concentration parameter of circular normal distribution

In this paper we discuss robust estimation of the concentration parameter (κ) of the circular normal (CN) distribution. It is known that the MLE of the concentration parameter is not B-robust at the family of all circular normal distributions with fixed mean direction (μ) and varying κ > 0. In this paper we propose a new estimator for κ and show that it is B-robust and SB-robust at the family {CN(μ, κ) : m ≤ κ ≤ M} where m and M are two arbitrary constants.     

Comparison of estimators in nested case–control studies with multiple outcomes

Reuse of controls in a nested case-control (NCC) study has not been considered feasible since the controls are matched to their respective cases. However, in the last decade or so, methods have been developed that break the matching and allow for analyses where the controls are no longer tied to their cases. These methods can be divided into two groups; weighted partial likelihood (WPL) methods and full maximum likelihood methods. The weights in the WPL can be estimated in different ways and four estimation procedures are discussed. In addition, we address modifications needed to accommodate left truncation. A full likelihood approach is also presented and we suggest an aggregation technique to decrease the computation time. Furthermore, we generalize calibration for case-cohort designs to NCC studies. We consider a competing risks situation and compare WPL, full likelihood and calibration through simulations and analyses on a real data example.     

Estimating detailed distributions from grouped sociodemographic data: ‘get me started in’ curve fitting using nonlinear models

In much demographic analysis, it is important to know how occurrence-exposure rates or transition probabilities vary continuously by age or by time. Often we have coarse or fluctuating data so there can be a need for estimation and smoothing. Since the distributions of rates or counts across age or another variable are often curved, a nonlinear model is likely to be appropriate. The main focus of this paper is on the estimation of detailed information from grouped data such as age and income bands; however, the methods we outline could also be applied to other settings such as smoothing rates where the original data are ragged. The ability to carry out curve fitting is a very useful skill for population geographers and demographers. Curve fitting is not well covered in statistics textbooks, and whilst there is a large literature in journals thoroughly discussing the detail of functions which define curves, these texts are likely to be inaccessible to researchers who are not specialists in mathematics. We aim here to make nonlinear modelling as accessible as possible. We demonstrate how to carry out nonlinear regression using SPSS, giving stepped-through hypothetical and research examples. We note other software in which nonlinear regression can be carried out, and outline alternative methods of curve fitting.     

Using remote sensing and census tract data to improve representation of population spatial distribution: case studies in the Brazilian Amazon

This work proposes a methodological approach to redistribute population data obtained from polygonal census tracts into population density surfaces (grids) based on a cell space database. The methodology was first developed for the municipality of Marabá, Pará state, in the Brazilian Amazon. We used a dasymetric method to eliminate areas of environmental restriction to human presence; then integrated environmental data indicative of human presence to generate a potential surface of population occurrence; and finally, census population count data were redistributed into cells. The methodology was subsequently adapted for 13 municipalities of the Sustainable Forests District (SFD) of BR-163, generating population distribution surfaces for 2000 and 2007. The evolution of the resident population over the SFD-BR163 showed spatial patterns compatible with the occupation process described in the literature and verified by fieldwork. To be applied over other areas, the proposed methodology must be adapted with local parameters but in this way, population density surfaces can be useful as an additional data source to study population and environment relationships.     

AIDS, “reversal” of the demographic transition and economic development: evidence from Africa

Using country- and region-level data, I investigate the effect of HIV/AIDS on fertility in Africa during 1985–2000. Results differ depending on the variation used and the estimation method. Between estimates that exploit cross-sectional variation suggest a positive significant effect of HIV/AIDS on fertility, whereas within estimates that are identified of off time-series variation show both positive and negative results depending on the HIV/AIDS variable used. These within estimates are insignificant in most of the specifications.     

Variable selection in linear regression analysis with alternative Bayesian information criteria using differential evaluation algorithm

In statistical analysis, one of the most important subjects is to select relevant exploratory variables that perfectly explain the dependent variable. Variable selection methods are usually performed within regression analysis. Variable selection is implemented so as to minimize the information criteria (IC) in regression models. Information criteria directly affect the power of prediction and the estimation of selected models. There are numerous information criteria in literature such as Akaike Information Criteria (AIC) and Bayesian Information Criteria (BIC). These criteria are modified for to improve the performance of the selected models. BIC is extended with alternative modifications towards the usage of prior and information matrix. Information matrix-based BIC (IBIC) and scaled unit information prior BIC (SPBIC) are efficient criteria for this modification. In this article, we proposed a combination to perform variable selection via differential evolution (DE) algorithm for minimizing IBIC and SPBIC in linear regression analysis. We concluded that these alternative criteria are very useful for variable selection. We also illustrated the efficiency of this combination with various simulation and application studies.     

Energy performance evaluation of OECD countries using Bayesian stochastic frontier analysis and Bayesian network classifiers

More recently a large amount of interest has been devoted to the use of Bayesian methods for deriving parameter estimates of the stochastic frontier analysis. Bayesian stochastic frontier analysis (BSFA) seems to be a useful method to assess the efficiency in energy sector. However, BSFA results do not expose the multiple relationships between input and output variables and energy efficiency. This study proposes a framework to make inferences about BSFA efficiencies, recognizing the underlying relationships between variables and efficiency, using Bayesian network (BN) approach. BN classifiers are proposed as a method to analyze the results obtained from BSFA.