I am not aware of any such specific software.

I use STATA to compute EGI and other indices. My friend Jay Dev wrote one STATA ado file to compute EG index. You may add that ado in your STATA to compute EG index. link of that ado is given below:

http://fmwww.bc.edu/repec/bocode/e/egi.ado

Dear Pritam Datta,

Thank you for the information. Yes, that is what I already have. Thank you once again.

I'm not sure if the R software package 'dbmss' computes specifically the EGI, but have several functions to calculate point pattern indices to characterize the spatial concentration, communly used in economic geography and ecology: Topographic measures (as the Ripley's K); Relative measures (as the M index that compares two point pattern structures) and Absolute measures (Kd and Kemp that allows to tests for spatial randomness). The package documentation is available in: https://cran.r-project.org/web/packages/dbmss/vignettes/dbmss.pdf

Dear Caudillo Cos Camilo Alberto,

thank you for the paper.

Hello everyone, I also want to use the ADO file above, but I do not know how to instal it on stata. Is there anyone who can show me how? thank you in advance for your help.

Dear Rachid,

please use the following command to install our EGI ado,

ssc install egi

read relevant explanation from the help file. After you install the ado file just type the following line, help file will pop up.

help egi

Thanks and Regards,

Pritam

Dear Pritam,

Thank you very much for your answer, it has been very helpful to me. By the way I have another question if you allow. The values of the ellison glaeser index that I have are all negative! is it normal? I thank you for your responsiveness.

Dear Rachid,

EGI may take negative value for industries those are spatially diffusing.

Quite dear Pritam, but what I forgot to say is that these indiustries are concentrated from the point of view of the Hrfindhal and the Gini index.

Dear Rachid,

recall, Herfindahl and Gini index measure gross level of spatial concentration. Whereas EGI measure localization beyond that expected by pure randomness. Given the general distribution of manufacturing activity; negative values imply that they locate in the same area less often than one would expect (i.e. negative values measure plants choosing to locate more diffusely than expected by randomness).

Hi guys,

I have installed EGI on Stata, but I can't understand how it should be the structure of the data. Any help or step-by-step guide available?

Thanks in advance

Hello David,

I started using aggregated data at the beginning, but i realized that individual data must be used for each unit studied. For example, the company, its location and any information about it.

Tawhid Monzur You may please check https://openjournals.wu-wien.ac.at/ojs/index.php/region/article/view/267