Matlab - Normal equation for linear least squares model fitting

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
1 reply · 5K views
jmher0403
Messages
21
Reaction score
0

Homework Statement



Im trying to construct a function in MATLAB that models linear least squares model fitting through the use of normal equations.

Homework Equations



Normal equation

(A'*A*c)=(A'*y)

A= [column vector of all x ; column vector of all 1]
y= [column vector of all y]
c = [b;a] where a&b are coefficients of the best fit line

The Attempt at a Solution

function c=mregress(x,y)
A=[x 1]
B=[y]

c=(A'*A)\(A'*B);

Error in ==> mregress at 2
A=[x 1]

I am very new to MATLAB and don't know much about it.
I think the problem is representing x a column vector and 1 is also a column vector of 1.

How do I represent this?
 
Physics news on Phys.org
A=[x,ones(n,1)] where n is the number of elements in x.