Click here to see the SAS code.
Click here to see the example.

---------------------------------------------------------------

I found all the microblogs containing the special keywords that might
be related to the mystery epidemic outbreak:
(flu, fever, chill, sweat, ache, pain, fatigue, cough, breathing, nausea, vomit, diarrhea)
and then plotted their locations on the map, and animated it over time.
Each frame of the animation contains the current day, in addition to 
all the previous days (cumulative).

In this analysis, I'm looking for both geographical & temporal trends.
Of course, since the last 3 days of data had the lion's share of the
blogs-of-interest (which we found out from the bar chart), those 3 days 
get "crowded" and it's a little hard to determine anything specific
(especially since each map is cumulative).  I was hoping to see the
epidemic start in one location, and then "spread out" from there ...
but there was no really obvious geo-temporal trend like that uncovered
by this animation.

Therefore in the next map (mc1d) I plot just the blogs for the final
few days separately (no animation, and no cumulative effect).

Back to Samples Index