Calculate Confidence Interval after using mldivide for regression analysis

I'm trying to calculate the confidence interval of the coefficients of my second order fit. Here is my function:

a(v_a,v_r) = b1*v_a.^2 + b2*v_r + b3

It should be noted that I already have values for a, v_a and v_r from my experiments.

In order to calculate the coefficients I used the mldivide operator in Matlab, since I couldn't find another way to fit a function with two independent variables:

b = V\a

Where

V = [v_a.*v_a v_r ones(size(v_a,1))];

This gives me the coefficients which I can then use to fit my model (and determine other values in the function). However, I would like to determine the accuracy of the fit. I would like to determine it's 95% confidence interval and plot it. Matlab has a lot of functions to do this, but there doesn't seem to be a way to do it for the mldivide (\) operator. Does anyone know how I could do this?

Answer 1

[the cyclist]

Not answering your question directly, but ...

If you have the Statistics and Machine Learning Toolbox, then you could use fitlm or regress to estimate the coefficients.

[L]

I eventually ended up using a similar technique. Turns out this is the most efficient way to do this and I was just being stubborn.