Appendix
Index construction methodology
The construction of the Investment Attractiveness Index requires the combination of different data sources into one unified measure of investment attractiveness at the city level. Structurally, the Index is split into three pillars (Growth Potential, Local Skills, Local Infrastructure) which in turn consist of a varying number of indicators. To arrive at the indicator, and subsequent pillar and headline index rankings, the various data points need to be combined in order to produce the respective scores.
Scores for each city are generated based on its performance as measured by the particular indicator. For each indicator the same set of steps will be followed, allowing us to assign a value to each country:
- In order to account for outliers, each data point is checked to determine if it falls outside of the mean +/- 2 standard deviations (s.d.) range. If it does, the country is assigned a value equal to either mean + 2 s.d. or mean – 2 s.d.
- The min-max approach is used to assign an index value to each city. Specifically, the following formula is used:
- For the 2024 iteration of the Investment Attractiveness Index, the series minimums and maximums are dictated by the series minimum and maximums from the corresponding indicator in the 2023 iteration of the Index, in order to allow for comparability across the two iterations.
- Once scores are assigned to each city for each indicator based on the previous steps, the indicators are weighted to calculate the overall pillar score. These will then in turn be aggregated into the overall index score.
- The adjustment in the methodology used to construct the index, aimed at facilitating year-on-year comparisons, will change the interpretation of the scores at the extremes of the scale. Specifically, the scores of '0' (previously indicating the weakest performer) and '100' (previously indicating the strongest performer) will no longer hold the same significance, as scores can now go below 0 or above 100.
For the Investment Attractiveness Index, we employed an equally weighted approach across each indicator (for pillar scores) and each pillar (for the overall score). This approach provides a unique city score for each metric which allows us to present separate figures for each indicator and pillar as well as an overall city score.
The sources for each of the indicators are as follows: