AMS210 Homework 7

4.4:3,4,6,12,13,14 4.5:2,8,11,18 DUE: Friday Oct. 27, Beginning of Class

Solutions to Problems:


4.4:3) x =  3 [ 1 ]  +  0 [ 5 ]  +  (-1) [ 4 ]  =  [-1 ]
	      [-4 ]	  [ 2 ]		 [-7 ]     [-5 ]
 	      [ 3 ]	  [-2 ]		 [ 0 ]     [ 9 ]


4.4:4) x = -4 [-1 ]  +  8 [ 3 ]  +  (-7) [ 4 ]  =  [ 4+24-28 ]  =  [ 0 ]
	      [ 2 ]	  [-5 ]		 [-7 ]     [-8-40+49 ]     [ 1 ]
 	      [ 0 ]	  [ 2 ]		 [ 3 ]     [ 0+16-21 ]     [-5 ]


4.4:6)  Solve the system c_1 b_1 + c_2 b_2 = x for (c_1,c_2), 
 represented by the matrix

	[ b_1  b_2  x ]  =  [ 1   5   4 ]     [ 1   5   4 ]
			    [-2  -6   0 ]  ~  [ 0   4   8 ]

Thus c_2 = 8/4 = 2,  and c_1 = 4 - 5c_2 = -6,
so [x]_B (the representation of x in the basis B) is  [ c_1 ]  =  [-6 ]
						      [ c_2 ]     [ 2 ]


4.4:12) Find [x]_B USING AN INVERSE matrix.   Recall x = (P_B)([x]_B)
where P_B is the matrix having columns the vectors in the ordered basis B.

	 P_B  =  [ 4   6 ]
	         [ 5   7 ]

  Thus  (P_B)^-1  =  1/(28-30) [ 7  -6 ]  =  -1/2 [ 7  -6 ]  =  [-7/2   3 ]
			       [-5   4 ]	  [-5   4 ]	[ 5/2  -2 ]

Alternatively the general procedure to find an inverse when we don't have a 
formula (as we did above for 2x2 matrix) is to adjoin P_B to I (the identity)
and row reduce P_B to I to get the matrix [ I   (P_B)^-1 ].

	[  4   6   1   0 ]             [  1  3/2  1/4   0  ]
	[  5   7   0   1 ]      ~      [  1  7/5   0   1/5 ]    ~
   
	[  1   3/2   1/4   0  ]         [  1   0  -7/2   3 ]  =  [ I   P_B^-1 ]
	[  0 -1/10  -1/4  1/5 ]    ~    [  0   1   5/2  -2 ]     
   
Now [x]_B = (P_B^-1) x  =  [-7/2   3 ] [ 2 ]  =  [-7 ].
			   [ 5/2  -2 ] [ 0 ]     [ 5 ]   


4.4:13) [p]_B = (c_1,c_2,c_3), where

c_1 (1 + t^2) + c_2 (t + t^2) + c_3 (1 + 2t + 2t^2) = p(t) = 1 + 4t + 7t^2

Equating like powers of t gives 
	c_1       +  c_3 = 1
	      c_2 + 2c_3 = 4
	c_1 + c_2 +  c_3 = 7

The augmented matrix for above system is 

	[  1   0   1   1 ]             [  1   0   1   1 ]
	[  0   1   2   4 ]      ~      [  0   1   2   4 ]    ~
	[  1   1   1   7 ]             [  0   1   0   6 ]

	[  1   0   1   1 ]             [  1   0   0   2 ]
	[  0   1   2   4 ]      ~      [  0   1   0   6 ]    
	[  0   0  -2   2 ]             [  0   0   1  -1 ]

Thus [p]_B = [ 2 ] , the representation of p relative to the basis B.
	     [ 6 ]
	     [-1 ]


4.4:14) Again let [p]_B = (c_1,c_2,c_3), where

c_1 (1 - t^2) + c_2 (t - t^2) + c_3 (2 - 2t + t^2) = p(t) = 3 + t - 6t^2

Equating like powers of t gives 
	c_1       + 2c_3 =  3
	      c_2 - 2c_3 =  1
       -c_1 - c_2 +  c_3 = -6
Note this is the same system as [p]_S = P_B [p]_B, where S is the standard 
basis for the space of all polynomials of degree <= 2, that is S = {1,t,t^2}.

The augmented matrix for the system is 

	[  1   0   2   3 ]             [  1   0   2   3 ]
	[  0   1  -2   1 ]      ~      [  0   1  -2   1 ]    ~
	[ -1  -1   1  -6 ]             [  0  -1   3  -3 ]

	[  1   0   2   3 ]             [  1   0   0   7 ]
	[  0   1  -2   1 ]      ~      [  0   1   0  -3 ]    
	[  0   0   1  -2 ]             [  0   0   1  -2 ]

Thus [p]_B = [ 7 ] , the representation of p in the basis B.
	     [-3 ]
	     [-2 ]


4.5:2)
a) V = {[  4s ] : s,t in R }  =   { s [ 4 ]  +  t [ 0 ] : s,t in R }
	[ -3s ]	             	      [-3 ]       [ 0 ]
	[  -t ]         	      [ 0 ]       [-1 ]

 Thus { [ 4 ]  [ 0 ] } spans the subspace, and note the vectors are linearly
	[-3 ], [ 0 ]
	[ 0 ]  [-1 ]
independent, so the set is a basis.

b) The dimension of the subspace is the # of vectors in the basis; that is
dim V = 2.


4.5:8) a)  V = { (a,b,c,d) : a-3b+c = 0 } = { (a,b,c,d) : c = -a+3b } = 

  =   { [  a  ] : a,b,d in R }  =  { a[ 1 ] + b[ 0 ] + d [ 0 ] : a,b,d in R }
	[  b  ]			      [ 0 ]    [ 1 ]     [ 0 ]
	[-a+3b]			      [-1 ]    [ 3 ]     [ 0 ]
	[  d  ]			      [ 0 ]    [ 0 ]     [ 1 ]

The three vectors are clearly linearly independent (the only way to get the 
zero vector is if a = b = d = 0) and span V.  
Thus a basis is { (1,0,-1,0) , (0,1,3,0) , (0,0,0,1) }.
Alternatively, we could write 

V  =  { [3b-c ] : b,c,d in R }  =  { b[ 3 ] + c[-1 ] + d [ 0 ] : b,c,d in R }
	[  b  ]			      [ 1 ]    [ 0 ]     [ 0 ]
	[  c  ]			      [ 0 ]    [ 1 ]     [ 0 ]
	[  d  ]			      [ 0 ]    [ 0 ]     [ 1 ]

so an alternative basis for V is { (3,1,0,0), (-1,0,1,0), (0,0,0,1) }.

b) dim V = 3.


4.5:11) First note that the 4 vectors are all in R^3, so the dimension of
span { v1, v2, v3, v4 } must be less than or equal to 3.
 
	 S = {  [ 1 ]  ,  [ 3 ]  ,  [ 9 ]  ,  [-7 ] } = { v1, v2, v3, v4 }
		[ 0 ]	  [ 1 ]	    [ 4 ]     [-3 ]
		[ 2 ]	  [ 1 ]	    [-2 ]     [ 1 ]

By the Spanning Set Thm., some subset of S is a basis for span S. We determine
a subset that is linearly independent.  Let 

  A  =  [ v1  v2  v3  v4 ]   =     [ 1   3   9  -7 ] 
			           [ 0   1   4  -3 ]	~
			           [ 2   1  -2   1 ]

	[  1   3   9  -7 ]             [  1   3   9  -7 ]
	[  0   1   4  -3 ]      ~      [  0   1   4  -3 ]    
	[  0  -5 -20  15 ]             [  0   0   0   0 ]

Col A is the same subspace as span S. Reduction to the echelon form above 
shows that Columns 1 and 2 are pivot columns. 

Hence dim(span S) = dim(Col A) = # of pivot columns of A =  2.

Note { v1, v2 } = { [ 1 ] , [ 3 ] } the set of (original) pivot columns of A, 
		    [ 0 ]   [ 1 ]	 is a basis for span S.
		    [ 2 ]   [ 1 ]

4.5:18)  We can see  A = [ 1   4  -1 ]  is already in an echelon form. A
			 [ 0   7   0 ]
			 [ 0   0   0 ]

has 2 pivot columns; also the # free variables for Ax = 0 equals the number of
nonpivot columns which is 1.
   Thus dim(Col A) = 2
	dim(Nul A) = 1.