Weighted Least Squares, Need Help!!

A question asks to determine write a function to solve the weighted least squares problem, without using advanced functions (limited to mldivide, for loops,etc). I know a weighted least squares problem can be solved by the basic least squares method after multiplying both Yi and the ith row of X by wi, but I'm not sure how to write any of the code in this case.

Answer 1

[Tom Lane]

Well, y.*w multiplies each element of y by the corresponding element of w. To do this with a matrix x having multiple columns, try "help repmat" or "help bsxfun". Also consider whether you need to use sqrt(w) or w -- it depends on how you define your weights.