Is it possible use the new 10.3 space time cube tools to visualise patterns in census block polygons?
Right now, the Space Time Pattern Mining tools are for use with incident data only, which are generally always points although there are certainly exceptions.
If you were looking at forest fires or flooding events, which are typically represented as polygons, you could use a centroid or point to represent the event/occurrence. However, boundaries such as census blocks are not true events and it probably wouldn't make sense. What we are probably interested in in that case is how the associated attribute is changing over time.
We are currently working on some very exciting ways to aggregate and analyze attributes in the cube structure, but they are not quite ready yet
Can you tell me more about the question that you are trying to answer? I could maybe suggest different way ...
Hope that helps,
Hi Jenora, thank you for your help.
I have 5 years of local government data stored at the census polygon level and I would like to see if there is a space/time relationship over time with certain demographic characteristics.
Retrieving data ...