The claim that Ward’s linkage algorithm in hierarchical clustering is limited to use with Euclidean distances is investigated. In this paper, Ward’s clustering algorithm is generalised to use with l1 norm or Manhattan distances. We argue that the generalisation of Ward’s linkage method to incorporate Manhattan distances is theoretically sound and provide an example of where this method outperforms the method using Euclidean distances. As an application, we perform statistical analyses on languages using methods normally applied to biology and genetic classification. We aim to quantify differences in character traits between languages and use a statistical language signature based on relative bi-gram (sequence of two letters) frequencies to calculate a distance matrix between 32 Indo-European languages. We then use Ward’s method of hierarchical clustering to classify the languages, using the Euclidean distance and the Manhattan distance. Results obtained from using the different distance metrics are compared to show that the Ward’s algorithm characteristic of minimising intra-cluster variation and maximising inter-cluster variation is not violated when using the Manhattan metric.
Various biophysical and socio-economic impacts may be associated with unconventional oil and gas (UOG) extraction. A vulnerability map may assist governments during environmental assessments, spatial planning and the regulation of UOG extraction, as well as decision-making around UOG extraction in fragile areas. A regional interactive vulnerability map was developed for UOG extraction in South Africa. This map covers groundwater, surface water, vegetation, socio-economics and seismicity as mapping themes, based on impacts that may emanate from UOG extraction. The mapping themes were developed using a normative approach, where expert input during the identification and classification of vulnerability indicators may increase the acceptability of the resultant map. This article describes the development of the interactive vulnerability map for South Africa, where UOG extraction is not yet allowed and where regulations are still being developed to manage this activity. The importance and policy implications of using vulnerability maps for managing UOG extraction impacts in countries where UOG extraction is planned are highlighted in this article.
The study uses an asset index of consumer durables to track changes in household wealth in Ghana during the recent period of strong growth. Using the Ghana Living Standards Survey of 1998, which contains both wealth data and consumer durable data, the chapter demonstrates that the asset index approximate marketable wealth adequately. Although asset index estimates of wealth cannot match the precision of wealth surveys, this approach can provide useful information on marketable wealth in countries where more appropriate sources are not available. The asset index analysis with the three demographic and health surveys for 1993, 1998, and 2003 suggests that the solid economic growth seen over this period has been accompanied by a strong rise in the average asset index scores.
Following the discussions documented by [3], guidelines were devised to support active earlycareer,emerging doctoral supervisors in Statistics in South Africa. These guidelines developedare incorporated into a guiding rubric and this associated usage manual. The rubricis not presented as a formal set of rules, but rather a dynamic document encouraging thegrowth of both the novice supervisor and the doctoral student. The rubric need not be usedin its entirety either, since it is only intended to aid in the supervision process within thediscipline of Statistics; it is not meant to be overwhelming or overbearing in the supervisionprocess.We present the current evolution of this guiding rubric, after discussions with, and feedbackfrom, both novice and senior supervisors within South Africa over the period from 2022-2024.If the rubric is used, the current paper and the conference paper [14] should be cited.This document is compiled with the intention of improving the potential for emerging supervisorsin Statistics to be appointed as primary supervisor, while allowing for the inclusion ofsenior academics as expert co-supervisors or mentors in the team. While the developmentof the doctoral student is the primary goal of this guiding rubric, the development of thenovice supervisor is also important as a secondary goal, contributing to the sustainability ofacademia in Statistics.