When we say variance of a normal is 0 it is typical to just take the distribution to be a point mass at the mean; in your case, this is 0. The distribution is defined by P(X = 0) = 1, and obviously doesn't admit a density.
Normal distribution - Wikipedia, the free encyclopedia
http://upload.wikimedia.org/math/9/e...9a0b69fbf6.png
The very first formula you see, i'm having to use that.
Basically i'm writing a program for Naive Bayes Classifier.
One of the classes has a standard deviation of zero.
Now, i can't divide by Zero...so if something has a S.D is zero, i gather that all value's must equal the mean(which in my case annoyingly is also zero).
So, if the SD is zero, how can i use the formula there?