# Basic statistics help, regression

1. Feb 10, 2012

### mathpariah

1. The problem statement, all variables and given/known data
Ive two questions not directly relating to a problem but just stat in general.

i)
In Swedish what is called the "help variable" for a regression equation: y=u'*B where u' is the vector of explaining variables (x1,...,xn) can apparently be written in many ways..

One way is this:

($\bar{X}$-μ) / s*√(u$^{T}$(X$^{T}$X)$^{-1}$u)

OR

in some cases Ive found its written:

($\bar{X}$-μ) / s*√(1 + u$^{T}$(X$^{T}$X)$^{-1}$u)

ii)

is variance s=√(μ/n)?

2. Relevant equations

What is the difference? It obviously affects the interval when you calculate it whether or not you have a 1 there or not.

3. The attempt at a solution