confidence interval for least square fitting

Hi, i have a system of equations of the form Ax=b. with A and b known matrics, i calculates x using least square method. Now i want to calculate confidence interval on my estimated values in x. Any idea how should i proceed?

Answer 1

[Star Strider]

For the parameter confidence intervals in a linear problem, use the regress function. The ‘bint’ output will have them.

[Sumera Yamin]

Many thanks for pointing this out.

[Star Strider]

As always, my pleasure!

[Sumera Yamin]

hi, i have a followup question. if i solve a system 'Ax=b', with a= Nx4 matrix, x=4x1 vector, and b=NX1 vector, then i am getting the exact same internal range /error for 1st and 3rd entry in x, same is true for 2nd and 4th entry in y. What could be the possible reson for this? I calculate error as |b-bint| (only upper or lower limit subtracted from b to get the error).

[Star Strider]

I am not certain that I understand what you are asking.

The ‘bint’ output is returned as an (Nx2) matrix. Just use it as it exists. No processing is necessary for it.