Find Docs

Astronomical Applications Department, U.S. Naval Observatory - Newcomb Manual (Page 11)

Wodocs >> Science : Astronomy >> Astronomical Applications Department, U.S. Naval Observatory Newcomb Manual Page 11
Astronomical Applications Department, U.S. Naval Observatory - Newcomb Manual
Suppose we have a scalar function of
M parame-
ters,
. Then the variation of
f is
f (p
1
,
, p
M
)
(16)
Df
=
S
k
=
1
M
f
p
k
Dp
k
h f
p
$ Dp
Now suppose
N measurements of the parameters
are taken and then the function
f computed from
p
these observationally determined parameters. The
variance of
f is then
(17)
r
f
2
=
1
N
(
f
i
-
f )
2
i
Approximate
by from (16). Then
f
i
-
f
Df
(18)
r
f
2
=
1
N
(
Df )
2
i
=
1
N

f
p
$ Dp
2

i
=
1
N

S
k
=
1
M
f
p
k
2
Dp
k
2
+
S
j, k
j
!k
f
p
j
f
p
k
Dp
j
Dp
k

i
=
S
k
f
p
k
2
1
N
(
Dp
k
)
2
i
+
S
j, k
j
!k
f
p
j
f
p
k
1
N Dp
j
Dp
k i
h
S
k
f
p
k
2
r
k
2
+
S
j, k
j
!k
f
p
j
f
p
k
r
jk
2
If the observations are uncorrelated, the cross terms
(double sum) tend to cancel. Equation (18) describes
the propagation of errors.
Now consider our least squares parameter esti-
mation, eq. (13). The parameters are quantities de-
a
termined from
N measurements,
, with
[
y
i
]
corresponding measurement errors
.
In light of
r
i
(18), and with a slight abuse of vector notation, we
may write the parameter variance as
(19)
r
a
2
=

a
y
i
2
r
i
2

i
+

a
y
i
a
y
j
r
ij
2
j

i
We will assume that the observational errors are
individually uncorrelated,
. Then we
r
ij
=
0
i ! j
are left with
(20)
r
a
2
=

a
y
i
2
r
i
2

i
Now, from (13) and (8) we have
(21)
a
=
C $ S
=
C $
y
i
r
i
Z(t
i
)
i
Thus,
(22)
a
y
i
=
1
r
i
C $ Z(t
i
)
Hence, (20) becomes
(23)
r
a
2
=
(
C $ Z(t
i
))
2
i
The k
th
component of (23) may be written
(24)
r
a
k
2
=
[
C $ Z(t
i
)]
k
[
C $ Z(t
i
)]
k i
Since is symmetric,
C
(25)
r
a
k
2
=
[
C $ Z(t
i
)]
k
[
Z(t
i
)
$ C]
k i
Interchange the order of summations:
(26)
r
a
k
2
=
[
C $ Z(t
i
)
Z(t
i
)
i
$ C]
kk
Using (9), we have
(27)
r
a
k
2
=
[
C $ A $ C]
kk
=
[
C $ C
-
1
$ C]
kk
=
[
C]
kk
which completes the derivation of (14).
3.4. Nonlinear Parameter Estimation.
Consider the condition equations (5). If the
model function
is nonlinear in the parameters
y(t, a)
, then the simple separation, eq. (6), which led to
a
the easily-solved normal equations, (12), is no longer
possible. Our goal is to minimize (4) with respect to
variation of the model parameters, despite the nonlin-
ear dependence of the model function on the parame-
ters. To do this, we will adopt an iterative approach.
Let
, where are the true (or, more accu-
a h a~
+
Da
a~
rately, best) values of the parameters. Assume we
start with parameter values that are close to the best
values. We can then approximate (4) with a trun-
cated Taylor series:
(28)
x
2
(
a) l x
2
(
a~)
-
Da $ B
+
1
2
Da $ M $ Da
where
Chapter 3: Parameter Adjustment
D:\Newcomb\Documentation\NewcombManual.lwp
11 of 19
10:42pm April 23, 1997

<< Previous       1 .. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18       Next >>



Other Documents:
Data Flow Diagrams, participant list, vondrak, zacharias, curvpath, pm AAS poster, Gaussian Distribution, Linear Least Squares, Sensitivity Integral, 3 D plot fits, lyap2, NAO 150, Rotate Vector, lyapcalc, wedges, Scifull, Precession Memo, Precession Maple, Curves 3 D, NOVAS 2004 Overview, NOVAS 2006 Overview, thesis, Ranson,


WODocs | |
All rights reserved. wodocs.com 2009
WODocs


New Docs

Documents Category:
Arts (Design, Movies, Music, Radio, Television)
Automotive (Cars, Marine, Motorcycle, ATV, Snowmobiles)
Business (Biotechnology and Pharmaceuticals, Chemicals, Construction and Maintenance, Materials, Real Estate, Services)
Electronics (Computers, Motion Control, Power Supply)
Games (Board, Family, Party, Card, Construction, RC Toys)
Health (Animalm, Beauty, Healthcare, Medicine, Pharmacy, Surgery, Weight Loss)
Home (Accessories, Cooking, Decor and Design, Electrical, Family, Pets)
News (Newspapers, Sports, Television)
Recreation (Collecting, Hiking, Scouting, Survival, Travel)
Reference (Education, Libraries and Archives, Museums)
Science (Agriculture, Astronomy, Biology, Chemistry)
Shopping (Antiques and Collectibles, Clothing, Flowers, Food, Home and Garden)
Sports (Bicycle, Snowboard, Skiing, Other)