To determine which of the given options are correct, we analyze each statement individually using matrix algebra, properties of determinants, and trace of matrices.
Step 1: Evaluation of Option (A)
We are given the matrices:
P = [ 1 0 0 0 0 1 0 1 0 ] , E = [ 1 2 3 2 3 4 8 13 18 ] , F = [ 1 3 2 8 18 13 2 4 3 ] P = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 1 & 0 \end{bmatrix}, \quad E = \begin{bmatrix} 1 & 2 & 3 \\ 2 & 3 & 4 \\ 8 & 13 & 18 \end{bmatrix}, \quad F = \begin{bmatrix} 1 & 3 & 2 \\ 8 & 18 & 13 \\ 2 & 4 & 3 \end{bmatrix} P = 1 0 0 0 0 1 0 1 0 , E = 1 2 8 2 3 13 3 4 18 , F = 1 8 2 3 18 4 2 13 3
First, let's calculate P 2 P^2 P 2 :
P 2 = [ 1 0 0 0 0 1 0 1 0 ] [ 1 0 0 0 0 1 0 1 0 ] = [ 1 0 0 0 1 0 0 0 1 ] = I P^2 = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 1 & 0 \end{bmatrix} \begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 1 & 0 \end{bmatrix} = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & 1 \end{bmatrix} = I P 2 = 1 0 0 0 0 1 0 1 0 1 0 0 0 0 1 0 1 0 = 1 0 0 0 1 0 0 0 1 = I
Next, we calculate P E P PEP P E P :
P E = [ 1 0 0 0 0 1 0 1 0 ] [ 1 2 3 2 3 4 8 13 18 ] = [ 1 2 3 8 13 18 2 3 4 ] PE = \begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 1 & 0 \end{bmatrix} \begin{bmatrix} 1 & 2 & 3 \\ 2 & 3 & 4 \\ 8 & 13 & 18 \end{bmatrix} = \begin{bmatrix} 1 & 2 & 3 \\ 8 & 13 & 18 \\ 2 & 3 & 4 \end{bmatrix} P E = 1 0 0 0 0 1 0 1 0 1 2 8 2 3 13 3 4 18 = 1 8 2 2 13 3 3 18 4
Multiplying P E PE P E by P P P on the right:
P E P = ( P E ) P = [ 1 2 3 8 13 18 2 3 4 ] [ 1 0 0 0 0 1 0 1 0 ] = [ 1 3 2 8 18 13 2 4 3 ] = F PEP = (PE)P = \begin{bmatrix} 1 & 2 & 3 \\ 8 & 13 & 18 \\ 2 & 3 & 4 \end{bmatrix} \begin{bmatrix} 1 & 0 & 0 \\ 0 & 0 & 1 \\ 0 & 1 & 0 \end{bmatrix} = \begin{bmatrix} 1 & 3 & 2 \\ 8 & 18 & 13 \\ 2 & 4 & 3 \end{bmatrix} = F P E P = ( P E ) P = 1 8 2 2 13 3 3 18 4 1 0 0 0 0 1 0 1 0 = 1 8 2 3 18 4 2 13 3 = F
Thus, F = P E P F = PEP F = P E P and P 2 = I P^2 = I P 2 = I .
Statement (A) is TRUE.
Step 2: Evaluation of Option (B)
First, let's compute the determinant of matrix E E E :
∣ E ∣ = ∣ 1 2 3 2 3 4 8 13 18 ∣ |E| = \begin{vmatrix} 1 & 2 & 3 \\ 2 & 3 & 4 \\ 8 & 13 & 18 \end{vmatrix} ∣ E ∣ = 1 2 8 2 3 13 3 4 18
Using row operations R 2 → R 2 − 2 R 1 R_2 \to R_2 - 2R_1 R 2 → R 2 − 2 R 1 and R 3 → R 3 − 8 R 1 R_3 \to R_3 - 8R_1 R 3 → R 3 − 8 R 1 :
∣ E ∣ = ∣ 1 2 3 0 − 1 − 2 0 − 3 − 6 ∣ = 1 ⋅ ( ( − 1 ) ( − 6 ) − ( − 2 ) ( − 3 ) ) = 6 − 6 = 0 |E| = \begin{vmatrix} 1 & 2 & 3 \\ 0 & -1 & -2 \\ 0 & -3 & -6 \end{vmatrix} = 1 \cdot ((-1)(-6) - (-2)(-3)) = 6 - 6 = 0 ∣ E ∣ = 1 0 0 2 − 1 − 3 3 − 2 − 6 = 1 ⋅ (( − 1 ) ( − 6 ) − ( − 2 ) ( − 3 )) = 6 − 6 = 0
Since ∣ E ∣ = 0 |E| = 0 ∣ E ∣ = 0 :
∣ F ∣ = ∣ P E P ∣ = ∣ P ∣ ∣ E ∣ ∣ P ∣ = 0 |F| = |PEP| = |P||E||P| = 0 ∣ F ∣ = ∣ P E P ∣ = ∣ P ∣∣ E ∣∣ P ∣ = 0
Now, using F = P E P F = PEP F = P E P and P 2 = I P^2 = I P 2 = I , we get P F = P ( P E P ) = P 2 E P = E P PF = P(PEP) = P^2 EP = EP P F = P ( P E P ) = P 2 E P = E P .
Thus, for the given non-singular matrix Q Q Q :
E Q + P F Q − 1 = E Q + E P Q − 1 = E ( Q + P Q − 1 ) EQ + PFQ^{-1} = EQ + EPQ^{-1} = E(Q + PQ^{-1}) E Q + P F Q − 1 = E Q + E P Q − 1 = E ( Q + P Q − 1 )
Taking the determinant on both sides:
∣ E Q + P F Q − 1 ∣ = ∣ E ( Q + P Q − 1 ) ∣ = ∣ E ∣ ⋅ ∣ Q + P Q − 1 ∣ = 0 ⋅ ∣ Q + P Q − 1 ∣ = 0 |EQ + PFQ^{-1}| = |E(Q + PQ^{-1})| = |E| \cdot |Q + PQ^{-1}| = 0 \cdot |Q + PQ^{-1}| = 0 ∣ E Q + P F Q − 1 ∣ = ∣ E ( Q + P Q − 1 ) ∣ = ∣ E ∣ ⋅ ∣ Q + P Q − 1 ∣ = 0 ⋅ ∣ Q + P Q − 1 ∣ = 0
Also:
∣ E Q ∣ = ∣ E ∣ ∣ Q ∣ = 0 ⋅ ∣ Q ∣ = 0 |EQ| = |E||Q| = 0 \cdot |Q| = 0 ∣ E Q ∣ = ∣ E ∣∣ Q ∣ = 0 ⋅ ∣ Q ∣ = 0
∣ P F Q − 1 ∣ = ∣ P ∣ ∣ F ∣ ∣ Q − 1 ∣ = ∣ P ∣ ⋅ 0 ⋅ ∣ Q − 1 ∣ = 0 |PFQ^{-1}| = |P||F||Q^{-1}| = |P| \cdot 0 \cdot |Q^{-1}| = 0 ∣ P F Q − 1 ∣ = ∣ P ∣∣ F ∣∣ Q − 1 ∣ = ∣ P ∣ ⋅ 0 ⋅ ∣ Q − 1 ∣ = 0
Therefore:
∣ E Q + P F Q − 1 ∣ = 0 = 0 + 0 = ∣ E Q ∣ + ∣ P F Q − 1 ∣ |EQ + PFQ^{-1}| = 0 = 0 + 0 = |EQ| + |PFQ^{-1}| ∣ E Q + P F Q − 1 ∣ = 0 = 0 + 0 = ∣ E Q ∣ + ∣ P F Q − 1 ∣
Statement (B) is TRUE.
Step 3: Evaluation of Option (C)
Using the multiplicative property of determinants:
∣ ( E F ) 3 ∣ = ∣ E F ∣ 3 = ( ∣ E ∣ ∣ F ∣ ) 3 = ( 0 ⋅ 0 ) 3 = 0 |(EF)^3| = |EF|^3 = (|E||F|)^3 = (0 \cdot 0)^3 = 0 ∣ ( E F ) 3 ∣ = ∣ E F ∣ 3 = ( ∣ E ∣∣ F ∣ ) 3 = ( 0 ⋅ 0 ) 3 = 0
∣ E F ∣ 2 = ( ∣ E ∣ ∣ F ∣ ) 2 = ( 0 ⋅ 0 ) 2 = 0 |EF|^2 = (|E||F|)^2 = (0 \cdot 0)^2 = 0 ∣ E F ∣ 2 = ( ∣ E ∣∣ F ∣ ) 2 = ( 0 ⋅ 0 ) 2 = 0
Since 0 > 0 0 > 0 0 > 0 is mathematically false, ∣ ( E F ) 3 ∣ > ∣ E F ∣ 2 |(EF)^3| > |EF|^2 ∣ ( E F ) 3 ∣ > ∣ E F ∣ 2 does not hold.
Statement (C) is FALSE.
Step 4: Evaluation of Option (D)
The sum of diagonal entries of any square matrix M M M is defined as its trace, Tr ( M ) \text{Tr}(M) Tr ( M ) .
Using the properties of trace:
Tr ( A + B ) = Tr ( A ) + Tr ( B ) \text{Tr}(A + B) = \text{Tr}(A) + \text{Tr}(B) Tr ( A + B ) = Tr ( A ) + Tr ( B )
Tr ( A B ) = Tr ( B A ) ⟹ Tr ( P − 1 E P ) = Tr ( E P P − 1 ) = Tr ( E ) \text{Tr}(AB) = \text{Tr}(BA) \implies \text{Tr}(P^{-1}EP) = \text{Tr}(EPP^{-1}) = \text{Tr}(E) Tr ( A B ) = Tr ( B A ) ⟹ Tr ( P − 1 E P ) = Tr ( E P P − 1 ) = Tr ( E )
Evaluating the sum of the diagonal entries for both expressions:
Tr ( P − 1 E P + F ) = Tr ( P − 1 E P ) + Tr ( F ) = Tr ( E ) + Tr ( F ) \text{Tr}(P^{-1}EP + F) = \text{Tr}(P^{-1}EP) + \text{Tr}(F) = \text{Tr}(E) + \text{Tr}(F) Tr ( P − 1 E P + F ) = Tr ( P − 1 E P ) + Tr ( F ) = Tr ( E ) + Tr ( F )
Tr ( E + P − 1 F P ) = Tr ( E ) + Tr ( P − 1 F P ) = Tr ( E ) + Tr ( F ) \text{Tr}(E + P^{-1}FP) = \text{Tr}(E) + \text{Tr}(P^{-1}FP) = \text{Tr}(E) + \text{Tr}(F) Tr ( E + P − 1 F P ) = Tr ( E ) + Tr ( P − 1 F P ) = Tr ( E ) + Tr ( F )
Since both expressions evaluate to Tr ( E ) + Tr ( F ) \text{Tr}(E) + \text{Tr}(F) Tr ( E ) + Tr ( F ) , they are equal.
Statement (D) is TRUE.
Conclusion
The correct options are A, B, and D .