next up previous
Next: A.1.6 Pull Functions Up: A.1 Basic Fitting Algorithm Previous: A.1.4 Convergence

A.1.5 Error Propagation

The final error matrices are defined by

Noting that , , and are independent of and , differentiate (A.13) with respect to and to get

Thus, using (A.27), (A.26), and (A.20),

(Note that since , the form can be written as , or .)

So, to evaluate , note that

and similarly for . So, using (A.21), one gets

 

Also, using (A.23),

Finally, using (A.22),

 



Scott Snyder Fri May 19 19:19:46 CDT 1995