← 2019 Paper 1
UPSC 2019 Maths Optional Paper 1 Q4a — Step-by-Step Solution 15 marks · Section A
Cayley-Hamilton theorem · Linear Algebra · asked 3× in 13 yrs · Read the full method →
Question
State the Cayley-Hamilton theorem. Use this theorem to find A 100 A^{100} A 100 , where A = [ 1 0 0 1 0 1 0 1 0 ] A=\begin{bmatrix}1&0&0\\1&0&1\\0&1&0\end{bmatrix} A = 1 1 0 0 0 1 0 1 0 .
Technique
State CH; get p ( λ ) = ( λ − 1 ) 2 ( λ + 1 ) p(\lambda)=(\lambda-1)^2(\lambda+1) p ( λ ) = ( λ − 1 ) 2 ( λ + 1 ) ; write λ 100 ≡ a λ 2 + b λ + c ( m o d p ) \lambda^{100}\equiv a\lambda^2+b\lambda+c\pmod{p} λ 100 ≡ a λ 2 + bλ + c ( mod p ) and pin a , b , c a,b,c a , b , c using the (repeated) eigenvalues, including the derivative condition at the double root.
Solution
Step 1 — Statement of the Cayley–Hamilton theorem
Cayley–Hamilton theorem. Every square matrix satisfies its own characteristic equation. If A A A is n × n n\times n n × n and p ( λ ) = det ( λ I − A ) p(\lambda)=\det(\lambda I-A) p ( λ ) = det ( λ I − A ) is its characteristic polynomial, then p ( A ) = O p(A)=O p ( A ) = O (the zero matrix).
Step 2 — Characteristic polynomial of A A A
p ( λ ) = det ( λ I − A ) = det [ λ − 1 0 0 − 1 λ − 1 0 − 1 λ ] = ( λ − 1 ) ( λ 2 − 1 ) = ( λ − 1 ) 2 ( λ + 1 ) . p(\lambda)=\det(\lambda I-A)=\det\begin{bmatrix}\lambda-1&0&0\\-1&\lambda&-1\\0&-1&\lambda\end{bmatrix}=(\lambda-1)\big(\lambda^2-1\big)=(\lambda-1)^2(\lambda+1). p ( λ ) = det ( λ I − A ) = det λ − 1 − 1 0 0 λ − 1 0 − 1 λ = ( λ − 1 ) ( λ 2 − 1 ) = ( λ − 1 ) 2 ( λ + 1 ) .
Expanded: p ( λ ) = λ 3 − λ 2 − λ + 1 p(\lambda)=\lambda^3-\lambda^2-\lambda+1 p ( λ ) = λ 3 − λ 2 − λ + 1 . By Cayley–Hamilton,
A 3 − A 2 − A + I = O ⟹ A 3 = A 2 + A − I . A^3-A^2-A+I=O\quad\Longrightarrow\quad A^3=A^2+A-I. A 3 − A 2 − A + I = O ⟹ A 3 = A 2 + A − I .
Step 3 — Reduce A 100 A^{100} A 100 via the characteristic polynomial
Divide λ 100 \lambda^{100} λ 100 by p ( λ ) = ( λ − 1 ) 2 ( λ + 1 ) p(\lambda)=(\lambda-1)^2(\lambda+1) p ( λ ) = ( λ − 1 ) 2 ( λ + 1 ) :
λ 100 = q ( λ ) ( λ − 1 ) 2 ( λ + 1 ) + ( a λ 2 + b λ + c ) . \lambda^{100}=q(\lambda)\,(\lambda-1)^2(\lambda+1)+\big(a\lambda^2+b\lambda+c\big). λ 100 = q ( λ ) ( λ − 1 ) 2 ( λ + 1 ) + ( a λ 2 + bλ + c ) .
Since p ( A ) = O p(A)=O p ( A ) = O , we will get A 100 = a A 2 + b A + c I A^{100}=aA^2+bA+cI A 100 = a A 2 + b A + c I . Determine a , b , c a,b,c a , b , c from the roots of p p p :
λ = 1 \lambda=1 λ = 1 (double root): substitute λ = 1 \lambda=1 λ = 1 :
1 = a + b + c . (i) 1=a+b+c. \tag{i} 1 = a + b + c . ( i )
Differentiate and put λ = 1 \lambda=1 λ = 1 (double root forces the derivative to match): 100 λ 99 = q ′ ( ⋅ ) p + q p ′ + ( 2 a λ + b ) 100\lambda^{99}=q'(\cdot)\,p+q\,p'+(2a\lambda+b) 100 λ 99 = q ′ ( ⋅ ) p + q p ′ + ( 2 aλ + b ) , and p ( 1 ) = p ′ ( 1 ) = 0 p(1)=p'(1)=0 p ( 1 ) = p ′ ( 1 ) = 0 , so
100 = 2 a + b . (ii) 100=2a+b. \tag{ii} 100 = 2 a + b . ( ii )
( − 1 ) 100 = 1 = a − b + c . (iii) (-1)^{100}=1=a-b+c. \tag{iii} ( − 1 ) 100 = 1 = a − b + c . ( iii )
From (i) and (iii): ( a + b + c ) − ( a − b + c ) = 2 b = 0 ⇒ b = 0 (a+b+c)-(a-b+c)=2b=0\Rightarrow b=0 ( a + b + c ) − ( a − b + c ) = 2 b = 0 ⇒ b = 0 . Then (ii) gives 2 a = 100 ⇒ a = 50 2a=100\Rightarrow a=50 2 a = 100 ⇒ a = 50 , and (i) gives c = 1 − 50 = − 49 c=1-50=-49 c = 1 − 50 = − 49 .
A 100 = 50 A 2 − 49 I . A^{100}=50A^2-49I. A 100 = 50 A 2 − 49 I .
Step 4 — Compute A 2 A^2 A 2 and assemble
A 2 = [ 1 0 0 1 0 1 0 1 0 ] 2 = [ 1 0 0 1 1 0 1 0 1 ] . A^2=\begin{bmatrix}1&0&0\\1&0&1\\0&1&0\end{bmatrix}^2=\begin{bmatrix}1&0&0\\1&1&0\\1&0&1\end{bmatrix}. A 2 = 1 1 0 0 0 1 0 1 0 2 = 1 1 1 0 1 0 0 0 1 .
Then
A 100 = 50 [ 1 0 0 1 1 0 1 0 1 ] − 49 [ 1 0 0 0 1 0 0 0 1 ] = [ 50 − 49 0 0 50 50 − 49 0 50 0 50 − 49 ] . A^{100}=50\begin{bmatrix}1&0&0\\1&1&0\\1&0&1\end{bmatrix}-49\begin{bmatrix}1&0&0\\0&1&0\\0&0&1\end{bmatrix}=\begin{bmatrix}50-49&0&0\\50&50-49&0\\50&0&50-49\end{bmatrix}. A 100 = 50 1 1 1 0 1 0 0 0 1 − 49 1 0 0 0 1 0 0 0 1 = 50 − 49 50 50 0 50 − 49 0 0 0 50 − 49 .
Answer
A 100 = [ 1 0 0 50 1 0 50 0 1 ] . \boxed{\;A^{100}=\begin{bmatrix}1&0&0\\50&1&0\\50&0&1\end{bmatrix}.\;} A 100 = 1 50 50 0 1 0 0 0 1 .