N - 1

Why do we use n-1?
n-1 is the degrees of freedom. Could be analogous to this, if we have a row of values ​​from a data, take the simple example of x1, x2, x3, and we also have to know the average (average = 100) from three data was then two of the three free valuable data. Suppose x1 = 150 and x2 = 50. From there, we can not determine x3 freely. So that's why on some analysis of the suspect who spoke only one parameter, independent degree is (n-1).

Degrees of freedom declared parts of the information contained in the free
data set, which is used to calculate a measure of the statistical basis. We know that
sum of the mean-corrected data is zero, and therefore the mean is also equal to zero. by
Hence the value of a mean-corrected that to - n can be determined from the number of
as (n-1)-corrected mean another. This means that there are only n - 1-corrected mean that
independent, or only n - 1 piece of information in the data mean-corrected. The reason that there are only n - 1 mean-corrected observations are independent is that m
observed was obtained by subtracting the mean of each observation
information is used to calculate the mean. Therefore degrees of freedom un
corrected is n - 1. Each basic measure that is calculated from the data sample mean
(Eg variance) will have a degree of freedom of n - 1.