Solutions to Projection and Least Square Problems, Study notes of Linear Algebra

Solutions to three projection and least square problems. The first problem involves finding the projection and residual of a vector on a subspace. The second problem deals with finding the least square linear fit to a given dataset. The third problem focuses on finding the least square solution to a system of linear equations. Each problem is solved using matrix algebra and the normal equations.

Typology: Study notes

Pre 2010

Uploaded on 07/29/2009

koofers-user-0u2
koofers-user-0u2 🇺🇸

9 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
1
Examples of projection/least square problems
Problem 1. Let S=R(A), where
A=
14
13
23
and b=
1
2
2
.
Find the projection pand residual rof bon S.
Solution. We have that
ATA=6 1
1 34and ATb=3
8.
Solving the linear system ATAc =ATbin cwe obtain c= [110/203,51/203]T. Hence
p=Ac = [94/203,43/203,373/203]T
and
r=bp= [297/203,363/203,33/203]T.
Problem 2. Find the least square linear fit to the data
x1 0 1 2
y21 0 3 .
Solution. Let
A=
11
1 0
1 1
1 2
and b=
2
1
0
3
.
We have that
ATA=4 2
2 6and ATb=0
8.
Solving the linear system ATAc =ATbin cwe obtain c= [4/5,8/5]T. Hence
y= (4/5)x+ 8/5
is the least square linear fit to the data.
pf2

Partial preview of the text

Download Solutions to Projection and Least Square Problems and more Study notes Linear Algebra in PDF only on Docsity!

Examples of projection/least square problems

Problem 1. Let S = R(A), where

A =

 (^) and b =

Find the projection p and residual r of b on S.

Solution. We have that

AT^ A =

[

]

and AT^ b =

[

]

Solving the linear system AT^ Ac = AT^ b in c we obtain c = [110/ 203 , − 51 /203]T^. Hence

p = Ac = [94/ 203 , 43 / 203 , 373 /203]T

and r = b − p = [− 297 / 203 , 363 / 203 , 33 /203]T^.

Problem 2. Find the least square linear fit to the data

x − 1 0 1 2 y − 2 − 1 0 3.

Solution. Let

A =

 and b =

We have that

AT^ A =

[

]

and AT^ b =

[

]

Solving the linear system AT^ Ac = AT^ b in c we obtain c = [− 4 / 5 , 8 /5]T^. Hence

y = (− 4 /5)x + 8/ 5

is the least square linear fit to the data.

Problem 3. Let

A =

 (^) and b =

Find the least square solution to the system Ax = b.

Solution. We have that

AT^ A =

[

]

and AT^ b =

[

]

Solving the linear system AT^ Ax = AT^ b in x we obtain that

x = [1/ 2 , 1 /2]T

is the least square solution to Ax = b.