uni algebra (matrics) (1 Viewer)

...

^___^
Joined
May 21, 2003
Messages
7,723
Location
somewhere inside E6A
Gender
Male
HSC
1998
lol..sorry, another one :(

Consider the system of linear equations

x<sub>1</sub> - x<sub>2</sub> + x<sub>3</sub> = 1,
2x<sub>1</sub> +2x<sub>2</sub> + x<sub>3</sub> = 5+1,
x<sub>1</sub> + 3x<sub>2</sub> - x<sub>3</sub> = 4+2,
x<sub>1</sub> + 7x<sub>2</sub> - 2x<sub>3</sub> = 10-1

a) reduce its associated augmented matrix to row-echolen form
b) fine a value of
c) solve the system


its too tricky for me brain! :(

edit: that λ is suppose to be that gamma symbol..lol
 
Last edited:

martin

Mathemagician
Joined
Oct 15, 2002
Messages
75
Location
Brisbane
Gender
Male
HSC
2002
x1 - x2 + x3 = 1,
2x1 +2x2 + x3 = 5+1,
x1 + 3x2 - x3 = 4+2,
x1 + 7x2 - 2x3 = 10-1

augmented matrix is (using L for lambda)
(You can imagine a dotted line between 3rd and 4th columns if you want to seperate the coefficents on LHS from constants on RHS)

1, -1, L, 1
2, 2, 1, 5L+1
1, 3, -1, 4L+2
1, 7, -2L, 10L-1

now we can add or subtract multiples of rows so doing Gaussian elimination
R2 -> R2 - 2R1
R3 -> R3 - R1
R4 -> R4 - R1

1, -1, L, 1
0, 4, 1-2L, 5L-1
0, 4, -1-L, 4L+1
0, 8, -3L, 10L-2

then
R3 -> R3 - R2
R4 -> R4 - 2R2

1, -1, L, 1
0, 4, 1-2L, 5L-1
0, 0, -2+L, -L+2
0, 0, -2+L, 0

I think this is row echelon form (reduced row echelon form is when you go all the way back up, right?)


thanks,
Martin
 
Last edited:

martin

Mathemagician
Joined
Oct 15, 2002
Messages
75
Location
Brisbane
Gender
Male
HSC
2002
For part 2 write out system in full
x1 - x2 + Lx3 = 1
4x2+(1-2L)x3 = 5L-1
(-2+L)x3 = -L+2
(-2+L)x3 = 0

so -2+L=0
therefore L=2

so

x1-x2+2x3=1
4x2-3x3=9
0=0
0=0

now this isn't enough info to solve fully but if we just think of x3 as a real parameter ie let x3=t

x2=1/4(9+3t)
x1 = 1+x2-2x3 = 1 + 1/4(9+3t) - 2t
= 13/4 - 5/4 t


so my solution (after about a million mistakes) is
x1 = 13/4 - 5/4 t
x2 = 1/4 (9+3t)
x3 = t
L = 2

we'll check say t=1 so
x1 = 2, x2 = 3, x3 = 1, L=2

this satisfies all the equations so it seems right.

thanks,
Martin
 
Last edited:

...

^___^
Joined
May 21, 2003
Messages
7,723
Location
somewhere inside E6A
Gender
Male
HSC
1998
Originally posted by martin
now we can add or subtract multiples of rows so doing Gaussian elimination
R2 -> R2 - 2R1
R3 -> R3 - R1
R4 -> R4 - R1

1, -1, L, 1
0, 4, 1-2L, 5L-1
0, 4, -1-L, 4L+1
0, 8, -3L, 10L-2

erk..how did u get -3L and 10L-2??

i got -2-L and 10L instead
 

martin

Mathemagician
Joined
Oct 15, 2002
Messages
75
Location
Brisbane
Gender
Male
HSC
2002
erk..how did u get -3L and 10L-2??
I had a mistake in the augmented matrix that I'd fixed later but forgotten to fix at the start (it should be right now)

-2L-L = -3L
10L-1-1 = 10L-2
 

...

^___^
Joined
May 21, 2003
Messages
7,723
Location
somewhere inside E6A
Gender
Male
HSC
1998
Originally posted by martin
I had a mistake in the augmented matrix that I'd fixed later but forgotten to fix at the start (it should be right now)

-2L-L = -3L
10L-1-1 = 10L-2
ah, thx

ur a champ...
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
I did this for a friend in /UTS but i got Gamma = 4/3 for a unique soln..

OMG i hope i didnt fark it up!?!!!!

but for r4 i got
[0 8 -2- L 10L-2]

then using reduced row algorithms i get
r3 =[0 0 -2+L -L+2]
r4 =[0 0 -16+12L 0 ]

for a value of L where the matrix is solvable (.. it says soluble? .. 10ppm in water??? crap sooo geeky)

then the only real value that x3 can take is -1 (x3=-1)

so we can conclude that -16+12L = 0
then r4 is now a row of zeros..
L = 4/3
for a solvable equation set...
 
Last edited:

martin

Mathemagician
Joined
Oct 15, 2002
Messages
75
Location
Brisbane
Gender
Male
HSC
2002
but for r4 i got
[0 8 -2- L 10L-2]
I originally did the same as you but I'm fairly sure now that
r1 = [1 -1 L 1]
r4 = [1 7 -2L 10L-1]

and r4 -> r4 - r1
= [0 8 -3L 10L-2]
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
Originally posted by martin
I originally did the same as you but I'm fairly sure now that
r1 = [1 -1 L 1]
r4 = [1 7 -2L 10L-1]

and r4 -> r4 - r1
= [0 8 -3L 10L-2]
yer ure right.. geez i did it too fast :S
r3 = [0 0 - 12 8 - 4L ]
r4 = [0 0 -16-4L 0 ]

for 3 unknowns then L= -4
since the bottom row has to be a row of zeros..
 

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
abdooooo!!!! read down.. there were mistakes made..
esp r3 and r4

about reduce row echelon form (rref)
1 0 0
0 1 0
0 0 1
0 0 0
would b the rref of a 4 x 3 matrix..
 

...

^___^
Joined
May 21, 2003
Messages
7,723
Location
somewhere inside E6A
Gender
Male
HSC
1998
hmm..

i got another one i couldn't do

let B = 1 - 6
6 10 -

a) for which vaules(s) of is B invertible
b) If = -1, solve Bx = 0
c) If = -2, solve Bx = 0

hmm..i dunno what invertible means..lol...bloody past papers with no solutions :(:(
 

wogboy

Terminator
Joined
Sep 2, 2002
Messages
653
Location
Sydney
Gender
Male
HSC
2002
Definition: Matrix A is invertible if and only if there exists an a matrix B such that: AB = I (I is the identity matrix)

Therefore det(AB) = det(I),
so det(A) * det(B) = det(I) (since the determinant of a product = the product of the determinants)
so det(A) * det(B) = 1 (determinant of identity matrix is 1)
det(B) = 1/det(A)

you can see that if det(A) = 0, the matrix B does not exist so B is not invertible. So that leads us to this theorem here:

Matrix A is invertible if & only if det(A) =/= 0

-------------------------------------

So for part a), find the determinant of your matrix B, and set this to equal zero. Note that this will give you all the values of such that A is NOT invertible.

For part b) and c), if = -1 or = -2 lies in the range that will make B invertible then x must be the zero vector (think about why this is true). Otherwise if B is not invertible, there will be infinitely many solutions including the zero vector. Use Gaussian elimination to solve in this case.
 
Last edited:

+Po1ntDeXt3r+

Active Member
Joined
Oct 10, 2003
Messages
3,527
Gender
Undisclosed
HSC
2003
B=
a b
c d

by the definition of det(B) = ad-bc =
so for invertible det(B) = 0 as wogboy sed.

(1 - )(10-) - 36= 0

solve.. then thats ure answer
 

...

^___^
Joined
May 21, 2003
Messages
7,723
Location
somewhere inside E6A
Gender
Male
HSC
1998
for some reason i can do inverse matrix.. :(:(

like find ther inverse of

2,5,8,5
1,2,3,1
2,4,7,2
1,3,5,3

i think u have to make 1s digaonlly down right?
 

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
what you do is
reduce the following to REDUCED row echelon form:

2 5 8 5 1 0 0 0
1 2 3 1 0 1 0 0
2 4 7 2 0 0 1 0
1 3 5 3 0 0 0 1

then the matrix on the right is the inverse.

there are otehr ways to do it .. but I think this is the best one in this case and most others
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top