Hello all,
I happen to get a (legitimate) hold of a city budget for the (4) years:
2006,2007,2008,2009
The budget holds over 12,000 rows of budget sections with numbers being
Zero's positive and negatives.
I would like to find something "interesting" in this dataset.
I don't have a clear definition of what this "interesting" might
be, nor how
to find it. But my aim is to find where the city council did something
"fishy" (again, no clear definition).
My hope is to try and use the time element to catch "something" on the
variables.
My initial idea was to try to use each section 4 (time) data points, and
maybe check
1) correlations and clusters within the section. to find "suspicious
similar" sections.
2) Also, I was hoping to make a small model for each section, and see if it
had major 1 outlier relative to the other 3 data points it has. (I feel that
is serious stretching of the data though...)
I would love for any interesting ideas (analysis or visualization vise).
Best,
Tal
----------------------------------------------
Contact me: Tal.Galili@gmail.com | 972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | *
www.r-statistics.com*/ (English)
[[alternative HTML version deleted]]
