Question: Standard errors in LinearFit

Hello,

I'm trying to fit some of my research data to a polynomial.   I can do this, but what I need to know is how Maple is calculating the standard errors.  In other words. I need to know the underlying forumla Maple is using for these standard errors.

Is there any documentation on this anywhere?  

For reference, the Maple command I am using is:
LinearFit([Z^(-2),Z^(-3),Z^(-4),Z^(-5),Z^(-6),Z^(-7),Z^(-8),Z^(-9)],X,P2,Z,output=standarderrors); 

with

X:=Vector([2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18],datatype=float[8]);
P2 is another vector with the corresponding y-axis values for each X value.

What I'm hoping for is something similar to this document sheet from OriginLab http://www.originlab.com/www/helponline/origin/en/UserGuide/Polynomial_Regression_Results.html

Thanks in advance! I really appreciate any help.   

-Wren