Method Moran Scatter Plot for EB Rates

[Gislaine Miranda Quaglio]

Dear all

I'm working with city-level data and there's a lot of heterogeneity. The variables are the industry value added (IVA) and the total value added (TVA, of all sectors). My question is: for me to use the participation rate (IVA/TVA) for spatial autocorrelation (global and local), the best option would be using the method Moran Scatter Plot for EB Rates. 

I ask this because, due to spatial heterogeneity, I have municipalities (or cities) with very different added values, however, when calculating the participation rate they become similar. For example, in A the IVA is $100,000 and the TVA is $200,000, but in B the VAI is $2,000 and $4,000, so their gross values ​​are very different, but their participation rates will be the same, 50%.

Thank you.

Best regards.