A posteriori variance factor in Covariance matrix

[xstamat...@gmail.com]

Hi all,

As we know the Covariance matrix C = (J' S J)^{-1}, where S is a weight matrix.

In many textbooks it is mentioned that there is an a-posteriori variance factor sigma_0 (shortening it to s_0) or unit weight that is calculated by s_0^2 = (v' S v) / r, with v being the residuals and r being the degrees of freedom or redundancy with r = n-u+f,

where n = number of observations, u = number of unknowns, f = number of constraints.

Then covariance matrix is then Q = s_o^2 * C. 

Should not that be applied to the covariance values that are calculated by ceres so that the estimated covariances are scaled correctly?

Thanks in advance,

[Sameer Agarwal]

I believe numpy/scipy does something like this. You are welcome to apply such a scaling yourself to the output computed by Ceres.

Sameer

--
You received this message because you are subscribed to the Google Groups "Ceres Solver" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ceres-solver...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ceres-solver/c69684e1-2a92-48b0-9be9-7b3bdd05bc10n%40googlegroups.com.