Next:
Estimate of Errors
Up:
3: Estimation of
Previous:
Formulation of Linear
Solution and Errors from Linear
We can write the result in matrix form
We can find errors in either of the two ways
a) regard
as a function of
and calculate standard deviation from central limit theorem.
b) Use the shape of likelihood,
.
We have---as far as shape of
L
is concerned:
(independent of
no terms linear in
due to ``maximum'' condition
),
i.e.,
matrix
M
is indeed error matrix for
a
.
Geoffrey Fox
,
Northeast Parallel Architectures Center
at Syracuse University,
gcf@npac.syr.edu
a) regard 
as a function of 
and
calculate standard deviation from central limit theorem.
b) Use the shape of likelihood, 
.
(independent of 
no terms linear in
due to ``maximum'' condition
),
matrix M is indeed error matrix for a.