topic Space Time Cube Analysis without 10 time intervals in Spatial Statistics Questions

Space Time Cube Analysis without 10 time intervals

Patatak_11 — Thu, 16 May 2024 14:36:02 GMT

Good morning!

I'm working on an analysis of change over time for populations within an existing polygon shape file. I have 5 years of historical population data per relevant polygon. However, when I run the Create Space Time Cube from Defined Locations, I receive error 110004: A STC must have 10 time slices.

Is there a way around this?

I do not have access to 10 year's worth of data, only the 1- year interval population counts for 5 consecutive years. I considered interpolating my population to 6 month intervals, but those wouldn't necessarily be accurate and seems like false replication. I want to do an actual analysis on the population count's change over time, and so adding false data seems like a poor approach.

TIA!

Patrick

Re: Space Time Cube Analysis without 10 time intervals

EdwardGause — Fri, 17 May 2024 16:57:24 GMT

I believe if you do the 6 month intervals, what you would have to do is Interpolate the values between the start of a year and the end of the year, which I think doing the average of the start and end values would be fine for the half a year Mark. Doing that would only leave Bias at the half year mark of the first year or at the half year mark of the last year depending on how you align you Time Interval dimension. So, yes you would be creating records out of thin air for data you did not have at the Half Year mark, but since you are averaging the start and end, you will get a good estimate for the Half Year mark, and that would be more accurate guess for value instead of just duplicating the value for either the beginning of the year or the end of the year.