On the Inconsistency of the Change-Point Estimator for the NE Family

Here, we provide a simple proof of the well-known classical result that the estimator of an unknown change-point is inconsistent. The proof utilizes only the law of large numbers in place of the usual random walk theory based arguments     

Predictability and interpretability of hybrid link-level crash frequency models for urban arterials compared to cluster-based and general negative binomial regression models

Machine learning (ML) techniques have higher prediction accuracy compared to conventional statistical methods for crash frequency modelling. However, their black-box nature limits the interpretability. The objective of this research is to combine both ML and statistical methods to develop hybrid link-level crash frequency models with high predictability and interpretability. For this purpose, M5′ model trees method (M5′) is introduced and applied to classify the crash data and then calibrate a model for each homogenous class. The data for 1134 and 345 randomly selected links on urban arterials in the city of Charlotte, North Carolina was used to develop and validate models, respectively. The outputs from the hybrid approach are compared with the outputs from cluster-based negative binomial regression (NBR) and general NBR models. Findings indicate that M5' has high predictability and is very reliable to interpret the role of different attributes on crash frequency compared to other developed models.     

Poverty in South Africa: Statistical inference for decomposable poverty measures

The measurement of poverty involves two problems: identifying the poor and constructing an index to measure the intensity of poverty. The headcount ratio is one of the widely used indices for the measurement of poverty. This article considers some of the other decomposable (i.e. additively separable) poverty measures. The importance of this property (decomposability) and of subgroup consistency for a poverty measure is highlighted. Since most of these poverty measures are estimated on the basis of sample surveys, the statistical significance of the observed differences is tested using the distribution-free approach proposed by Kakwani (1993, Review of Economics and Statistics , 75: 632-40). The Income and Expenditure Survey conducted in 1995 by Statistics South Africa, then called Central Statistical Service, is used in the study.     

Financial Performance of Foreign and Domestic Owned Companies in India

Foreign direct investment (FDI) is attracted into transitional economies like India for reasons like capital, technology, international markets, and managerial skills. Extant literature has focused on studying the impact of FDI inflows on financial performance at an aggregate level ignoring the differences in state regulations in industries. Further, in the case of India, the literature is confined to analysis in specific years. In this study, we examined whether foreign ownership has any impact on financial performance of firms in India over a period of time. We have conducted a pooled cross section time series analysis of 102 Indian pharmaceutical firms for the period 1998–2005. For the analysis, we used a novel, robust method of handling missing data thus making available twenty % more information than is otherwise possible. Foreign ownership is found to have a positive and statistically significant impact on the financial performance of pharmaceutical companies in India.     

Poverty in South Africa – a statistical analysis

Poverty is one of the major challenges facing democratic South Africa. This article focuses on poverty in South Africa, using the Income and Expenditure Survey conducted in 1995 by Statistics South Africa (formerly the Central Statistical Service). In the first part, different approaches that can be followed in the measurement of poverty are discussed. In the second part, Sen's approach to the measurement of poverty and the Chi-square Automatic Interaction Detection (CHAID) technique have been used to analyse the above data. CHAID is used to explore the relationship between the poverty status of the household (ie poor or non-poor) and other household characteristics. These variables can then be used as predictors of poverty status.