China 
Africa 
Explore chapters and articles related to this topic
Sustainability Analytics and Decision Acumen
Published in Ram Ramanan, Introduction to Sustainability Analytics, 2018
Indexes are sets of indicators aggregated into a single composite measure. There is a process for converting indicators to indexes. First, a group of indicators, meaningful and relevant to the index, are identified. The metrics for these indicators have to be normalized for aggregation. This process involves transforming original value of all the select group of indicators into a more generic dimensionless value. The final step involves assigning relative weights against each other, and a weighted average value yields aggregate indexes such as the Dow Jones Sustainability Index. This information is then accessed by investors through information channels that provide environmental, social, and governance (ESG) data to investors, usually as a subscription service such as the Bloomberg terminal where ESG performance data is housed alongside more traditional financial information.
Objective Quality and Intelligibility Measures
Published in Philipos C. Loizou, Speech Enhancement, 2013
The NCM uses the whole utterance to compute the correlation between the clean and processed envelopes in each band. Measures, similar to NCM, which compute the correlation between clean and processed envelopes in relatively shorter intervals (ranging from 20 to 384 ms in duration) and averaging the resulting correlations over the whole utterance, were proposed in [95,114,118–121]. The averaged correlation, computed across frames or blocks of frames, is used as a measure for predicting SI. In [95,121], for instance, the (squared) correlation of 20-band excitation spectra was computed using 20 ms frames. In [119], the (squared) correlation of 32-band excitation spectra, which have been normalized by their rms energy, was computed using 25.6 ms frames. In [114], the correlations of excitation spectra were computed using 192-frame blocks, with 32 ms duration frames and 75% overlap. Combining the correlation-based measures with other measures has proven to improve the correlation of a composite measure [95,114].
Implementing Performance Measures
Published in Will Kaydos, Operational Performance Measurement, 2020
The level of performance indicated by the composite measure is going to be very dependent on the weights assigned to the individual components. This requires making judgments that may have to be based on limited data. However, if the weights assigned are reasonably correct, the relative importance of individual problems and the degree of improvement in performance are not going to be significantly distorted.
Performance index for public housing in East Malaysia
Published in Architectural Engineering and Design Management, 2022
Olga Postnikova, J. A. Bamgbade, N. H. Wong
The next step is examining the empirical relationships among the selected items. There are two types of relationships among the items: bivariate and multivariate relationships. Examining empirical relationships can be achieved through statistical techniques useful for the next step of item scoring and further combining into an index. After deciding if the scores will be divided equally or not, the composite measure was constructed via a composite index based on the desired range of scores. The last consideration is the index validation that can be either external or internal. Internal validation was conducted – through item analysis (item selection) – on cases with larger numbers of items (Babbie, 2015).