2016 Lec 6

6.
JORDAN NORMAL FORM
The goal of this chapter is to give a good description of linear trans-

formations when restricted to the invariant subspaces that occur in
the Primary Decomposition Theorem.
Let V be finite dimensional and T : V → V be a linear transforma-

tion. If T n = 0 for some n > 0 then T is called nilpotent.
Theorem 6.1: If T is nilpotent, then its minimal polynomial has the

form m T ( x ) = x m for some m and there exists a basis B of V such
that:  
0 ∗ 0
 .. .. 
 . . 
 with ∗ = 0, 1.
B [ T ]B = 

.. 
 . ∗ 
0 0
Proof: As T is nilpotent, T n = 0 for some n, and hence m T ( x )| x n .

Thus m T ( x ) = x m for some m.
To prove the main statement, we first make the following three
observations.
(1.)
0 < ker( T ) < ker( T 2 ) < ... < ker( T m−1 ) < ker( T m ) = V.
By the minimality of m these inclusions are indeed strict as ker( T k ) =

ker( T k+1 ) implies that ker( T k ) = ker( T k+s ) for all s ≥ 0. (Details are
left as an exercise.)
(2.) Let Bi be such that:
B̄i = {w + ker( T i−1 )|w ∈ Bi } is a basis for ker( T i )/ker( T i−1 ).
From Proposition 3.2 and the surrounding discussion it follows by

induction that [
B = Bi
i
is a basis for V. (Details are left as an exercise.)

(3.) Next we observe that
{ Tw + ker( T i−1 )|w ∈ Bi+1 }
1
2
is linearly independent in ker( T i )/ker( T i−1 ). For:
Σs ( as Tws + ker( T i−1 )) = ker( T i−1 ) ⇒ Σs as Tws = T (Σs as ws ) ∈ ker( T i−1 )

⇒ Σs as ws ∈ ker( T i )
⇒ Σs as ws + ker( T i ) = ker( T i )
⇒ as = 0 for all s by the definition of Bi+1 .
We are now ready to construct the desired basis B . Starting with

i = m + 1 and Em = Bm+1 = ∅, we can find inductively Ei so that
Bi = Ei q T (Bi+1 ) descends to a basis B̄i for ker( T i )/ker( T i−1 ). Then
[
B= Bi
i
{ T m−1 w, ..., Tw, w} ∪ · · · ∪
[ [
= {w}
w∈Em w∈E1
is a basis of V, and:
 
A1
..
B [ T ]B =  is block diagonal
 
.
As
with |Ei | = k i many Jordan blocks of size i × i

 
0 1 0
 .. .. 
 . . 
Ji =  .
 .. 
 . 1 
0 0
Corollary 6.2: Let V be finite dimensional and T : V → V be a linear

transformation. Assume m T ( x ) = ( x − λ)m for some m. Then, there
exists a basis B of V such that B [ T ]B is block diagonal with blocks of
the form:
 
λ 1 0
 .. .. 
 . . 
 and i ≤ m.
Ji (λ) := λI + Ji = 
 .. 
 . 1 
0 λ
Proof: T − λI is nilpotent with minimal polynomial x m . We may

apply Theorem 6.1. So there exists a basis B such that B [ T − λI ]B is
block diagonal with blocks Ji and hence
B [ T ]B = λI +B [ T − λI ]B
is of the desired form.

3
Theorem 6.3: Let V be finite dimensional and let T : V → V be a

linear map such with minimal polynomial
m T ( x ) = ( x − λ 1 ) m1 · · · ( x − λ r ) mr .
Then there exists a basis B of V such that B [ T ]B is block diagonal and

each diagonal block is of the form Jk (λi ) with k ≤ mi .
Note: If F is an algebraically closed field, such as C, then the mini-

mal polynomial will always split into a product such as in the theo-
rem.
Proof: By the Primary Decomposition Theorem 5.3,
V = ker( T − λ1 I )m1 ⊕ · · · ⊕ ker( T − λr I )mr .
Furthermore, T restricted to the i-th summand has minimal polyno-

mial ( x − λi )mi and hence Corollary 6.2 applies to give the desired
result.
Example 6.4: Let T : R3 → R3 be given by
−2 −1 1
 
A =  14 7 −7 
10 5 −5
A2 = 0 ⇒ m A ( x ) = x 2 ( as A 6= 0)
3
⇒ χ A (x) = x
We have 0 < ker( T ) < ker( T 2 ) = R3 with (ker( T ) = h(1, 0, 2)T , (0, 1, 1)T i)
and
ker( T 2 )/ ker( T ) = h(1, 0, 0)T + ker( T )i.
So
B2 = {(1, 0, 0)T }
B1 = T (B2 ) ∪ E1 = {(−2, 14, 10)T , (0, 1, 1)T }
and B = B2 ∪ B1 = {(−2, 14, 10)T , (1, 0, 0)T , (0, 1, 1)T }
Hence  
0 1 0
B [ T ]B =  0 0 0 .
0 0 0
Example 6.5: Let T : R3 → R3 be given by

 
3 0 1
A =  −1 1 −1  .
0 1 2
4
Then χ T ( x ) = det( A − xI ) = · · · = (2 − x )3 and m T ( x ) = ( x − 2)3 .

 
1 0 1
( A − 2I ) =  −1 −1 −1 
0 1 0
 
1 1 1
( A − 2I )2 =  0 0 0 
−1 −1 −1
We have
0 < ker( A − 2I ) < ker(( A − 2I )2 ) < ker(( A − 2I )3 ) = R3
and
B3 = {(1, 0, 0)T } as ( A − 2I )2 (1, 0, 0)T 6= 0

B2 = ( A − 2I )B3 = {(1, −1, 0)T }
B1 = ( A − 2I )B2 = {(1, 0, −1)T }
⇒ B = B1 ∪ B2 ∪ B3 = {(1, 0, 0)T , (1, −1, 0)T , (1, 0, −1)T }

Put  
1 1 1
P =  0 −1 0  .
−1 0 0
Then  
2 1 0
P−1 AP =  0 2 1  .
0 0 2
Proposition 6.6: Let

 
λ 1 0
 .. .. 
 . . 
Jk (λ) =  
 .. 
 . 1 
0 λ
be a k × k Jordan block and consider
vn := Jk (λ)vn−1 = ( Jk (λ))n v0 .
Then, with the notation vn = {v1n , · · · , vkn }, we have

k −i n k −i n n −1 k − i +1 n n −i k
v n = λ v0 + λ v0 +···+ λ v0 .
1 i
Proof: We proceed by induction on n. Note for n = 0 there is nothing

to prove. By definition
vkn−i = λvkn− i k − i +1
−1 + v n −1 .
5
Using the induction hypothesis and the binomial identity
n−1 n−1

n
+ =
j j−1 j
proves the lemma.

2016 Lec 6

Uploaded by

Copyright:

Available Formats

2016 Lec 6

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2016 Lec 6

Uploaded by

Copyright:

Available Formats

6.

JORDAN NORMAL FORM

The goal of this chapter is to give a good description of linear trans-

Let V be finite dimensional and T : V → V be a linear transforma-

Theorem 6.1: If T is nilpotent, then its minimal polynomial has the

Proof: As T is nilpotent, T n = 0 for some n, and hence m T ( x )| x n .

By the minimality of m these inclusions are indeed strict as ker( T k ) =

B̄i = {w + ker( T i−1 )|w ∈ Bi } is a basis for ker( T i )/ker( T i−1 ).

From Proposition 3.2 and the surrounding discussion it follows by

is a basis for V. (Details are left as an exercise.)

{ Tw + ker( T i−1 )|w ∈ Bi+1 }

is linearly independent in ker( T i )/ker( T i−1 ). For:

Σs ( as Tws + ker( T i−1 )) = ker( T i−1 ) ⇒ Σs as Tws = T (Σs as ws ) ∈ ker( T i−1 )

We are now ready to construct the desired basis B . Starting with

with |Ei | = k i many Jordan blocks of size i × i

Corollary 6.2: Let V be finite dimensional and T : V → V be a linear

Proof: T − λI is nilpotent with minimal polynomial x m . We may

is of the desired form.

Theorem 6.3: Let V be finite dimensional and let T : V → V be a

Then there exists a basis B of V such that B [ T ]B is block diagonal and

Note: If F is an algebraically closed field, such as C, then the mini-

V = ker( T − λ1 I )m1 ⊕ · · · ⊕ ker( T − λr I )mr .

Furthermore, T restricted to the i-th summand has minimal polyno-

Example 6.4: Let T : R3 → R3 be given by

Example 6.5: Let T : R3 → R3 be given by

Then χ T ( x ) = det( A − xI ) = · · · = (2 − x )3 and m T ( x ) = ( x − 2)3 .

0 < ker( A − 2I ) < ker(( A − 2I )2 ) < ker(( A − 2I )3 ) = R3

B3 = {(1, 0, 0)T } as ( A − 2I )2 (1, 0, 0)T 6= 0

⇒ B = B1 ∪ B2 ∪ B3 = {(1, 0, 0)T , (1, −1, 0)T , (1, 0, −1)T }

Proposition 6.6: Let

be a k × k Jordan block and consider

Then, with the notation vn = {v1n , · · · , vkn }, we have

Proof: We proceed by induction on n. Note for n = 0 there is nothing

Using the induction hypothesis and the binomial identity

proves the lemma.

You might also like