How do i get the parameter uncertainties from a guassian fit?

[wideofthemark]

Hi,

Was wondering if there is an easy way of getting the uncertainties/variances (or effectively the covariant matrix) in the parameters in a gaussian fit?  I guess that they are calculated but i can't find a vi that will return them.  - Or am i going to have to set up the gaussian fit as a 'general nonlinear fit'. 

I can do this but it seems like labview has dropped the ball here if they just aren't allowing you to do this easily - since as i say they MUST be calculating it to do Least squares or whatever....  Surely a fit isn't any use if you don't know how good the fit parameters is.

JP

[DSPGuy]

Check out the mathematics->curve fitting->advanced palette for the Gaussian Peak Fit Intervals.vi.  While this does not output the covariance matrix, it does use the covariance to compute the confidence intervals for the parameters and the best fit curve.

-Jim

[wideofthemark]

Hmmmm.  I guess you are right - but it seems to do it in some system calling function, and it doesn't list the covariance matrix as an output.  I guess there might be a function in the system that will return a covariant matrix.