[Openspace] confidence intervals for changes in morans I over time

[Ola Claësson]

Hi there!!
 
 
I have a data set for six time periods covering mortality data for an area. I am interested to see if the spatial dependence in mortality changes over time, and I have therefore calculated Moran's I for all six periods. (It is important to mention that the number of spatial units under observation in the different time periods is not constant due to missing data, so the number range between 223 and 301). I wonder if it would be reasonable to calculate confidence intervals to assess any change in the Moran's I, or would there be any underlying statistical problems with such an approach? I am thankful for any thoughts about it.  
 
 
Sincerely 
 
Ola