Need help understanding a least squares problem

In this problem, you will write a function named MyRegressor to with data with n given functions.The header of this function is

function A = MyRegressor(x, y, FHs)

Here x and y are two column vectors of the same size. The relation between each element in x and its corresponding element in y can be described by y = a1 f1(x) + a2 *f2(x) + an fn(x)

f1, f2, , fn are n functions and the handles to these functions are contained in the n1 cell array FHs. You need to solve for the coecient array A by the least squares regression.

what type of solution am I solving for? unique? exact unique?

size(A) 
rank(A) 
rank([A b])

Answer 1

[Matt J]

Your question and the notation it uses is unclear, so I'll have to give you some abstract hints. If you have a linear system of equations

 A*x=b

where A is MxN, and b is Mx1, the uniqueness of the solution is determined by rank(A).

Whether the solution is exact is related to

rank([A,b]) - rank(A)