I was reading this paper related to Bayesian spatial scan statistics where I came across the Kulldorff's scan statistics.
I have attached the screenshot of the paper. My objective is to find a location of potential disease outbreak.
I have a little doubt with the formula given in the paper. In the paper it is mentioned that the Fscore of any location is
My question is when they calculate the F score.
The number of sick people in current location(Cin) = 5 The baseline population is around(Bin) 50 The number of sick people outside current location(Cout) = 2245 The number of sick people outside current location(Bout) = 62475
Call = 5+2245 = 2250 Ball = 50+62475 = 62525
Now if I calculate the Fscore it comes out to be Nan. Is it the correct formula? Any insights.
In the next link I found something like this
Now even though I use this formula, the value is NaN. It is because like in the case of denominator the prob < 1 and I am raising it to a high power, so it comes out to be zero. So my question is, is it the problem with my data itself? If I plug it into the formula it will definitely come out to be like that.
