Step 1: Understanding the Problem: This question concerns the rank of a product of two non-square matrices. We'll utilize fundamental properties of matrix rank, focusing on its relationship to matrix dimensions and products.
Step 2: Key Concept: The core property for the rank of a matrix product \( AB \) is:
\[ \rho(AB) \le \min(\rho(A), \rho(B)) \]
Also, a matrix's rank cannot exceed its number of rows or columns.
For \(P_{m \times n}\), \( \rho(P) \le \min(m, n) = m \) (since \(m < n\)).
For \(Q_{n \times m}\), \( \rho(Q) \le \min(n, m) = m \).
Step 3: Analysis of Statements:
Let's evaluate each statement:
A. \( \rho(PQ) = n \):
The product \(PQ\) has dimensions \( m \times m \). Its maximum rank is \(m\). Given \( m < n \), \( \rho(PQ) \) cannot be \(n\).
Also, \( \rho(PQ) \le \min(\rho(P), \rho(Q)) \le \min(m, m) = m \). Since \(m < n\), \( \rho(PQ) \) must be less than \(n\). Therefore, \( \rho(PQ) = n \) is impossible.
B. \( \rho(QP) = m \):
The matrix product \(QP\) has dimensions \( n \times n \). \( \rho(QP) \le \min(\rho(Q), \rho(P)) \le m \). Thus, the rank is at most \(m\). Matrices \(P\) and \(Q\) can be constructed such that \( \rho(P) = \rho(Q) = m \) and \( \rho(QP) = m \). This statement is possible.
C. \( \rho(PQ) = m \):
\(PQ\) is an \(m \times m\) matrix. Its rank can be at most \(m\). It's possible to construct \(P\) and \(Q\) such that \( \rho(P) = \rho(Q) = m \) and \( \rho(PQ) = m \). This statement is possible.
D. \( \rho(QP) = \lfloor (m+n)/2 \rfloor \):
We know \( \rho(QP) \le m \). For this to hold, \( \lfloor (m+n)/2 \rfloor \le m \). Since \( m < n \), we can find cases where this is violated. For example, let \( m = 2 \) and \( n = 5 \). Then \( \rho(QP) = \lfloor (2+5)/2 \rfloor = 3 \), but \( \rho(QP) \le m = 2 \). This is a contradiction. Therefore, this statement is generally impossible.
Step 4: Conclusion:
Statements A and D describe scenarios that are not generally feasible. Thus, the correct option is (A).