Tests and confidence intervals for economic, scientometric and technological specialisation ratios

In economic and scientometric applied research, often so-called specialisation indices are used, the most established one among them being the Balassa index. These indices measure comparative strenghts or waeknesses as well as specialisation profiles of the observation units with respect to certain criteria. This allows to answer question like: Is Germany specialised in the export of motor vehicles? Unfortunately, little is known about statistical properties of these indicators, which makes valid inferencing a hard task. In this article we prove asymptotical normality for a large class of specialisation indices, provide asymptotic confidence intervals and demonstrate in an example how to obtain statistically sound results. We will also address the problem of normalisation of these indicators. All procedures proposed are provided in an add-on package for R statistical environment in order to quarantee fast dissemination of this methodology.