Applications of Matrices and Determinants - Study Notes
Unit summary
Unit 1 of Class 12 Business Mathematics and Statistics uses matrices and determinants to answer three practical questions: how many independent rows a matrix really has (its rank), whether a set of linear equations can be solved and how (consistency and Cramer's rule), and how market shares move from one period to the next (transition probability matrices).
1.1 Rank of a matrix
- The rank of a matrix \( A \), written \( \rho(A) \), is the order of the largest non-zero minor of \( A \).
- \( \rho(A) = r \) means: at least one minor of order \( r \) is non-zero, and every minor of order \( r+1 \) and higher is zero.
- \( \rho(A) \geq 0 \); for an \( m \times n \) matrix, \( \rho(A) \leq \min\{m, n\} \).
- The rank of a zero matrix is 0; the rank of a non-singular matrix of order \( n \times n \) is \( n \); so \( \rho(I_n) = n \).
Elementary transformations
- Interchange of two rows or columns: \( R_i \leftrightarrow R_j \), \( C_i \leftrightarrow C_j \).
- Multiplying every element of a row or column by a non-zero scalar: \( R_i \to kR_i \).
- Adding a scalar multiple of another row or column: \( R_i \to R_i + kR_j \), \( C_i \to C_i + kC_j \).
A row is never combined with a column. Matrices obtained from each other by a finite number of elementary transformations are equivalent, written \( A \sim B \), and have the same rank.
Echelon form (up to order 3 x 4)
- Every row with all entries zero lies below every non-zero row.
- The number of zeros before the first non-zero element in a row is less than the number of such zeros in the next row.
- The rank equals the number of non-zero rows in the echelon form.
Consistency of equations
- Every system of linear equations has either no solution, exactly one solution or infinitely many solutions.
- A system with at least one solution is consistent; otherwise it is inconsistent.
- For \( AX = B \), \( A \) is the coefficient matrix and \( [A, B] \) is the augmented matrix.
| Ranks (n unknowns) | Result |
|---|---|
| \( \rho(A) = \rho([A, B]) = n \) | Consistent, unique solution |
| \( \rho(A) = \rho([A, B]) < n \) | Consistent, infinitely many solutions |
| \( \rho(A) \neq \rho([A, B]) \) | Inconsistent, no solution |
For a square system, \( |A| \neq 0 \) means \( A \) is non-singular and the system has a unique solution.
1.2 Cramer's rule
- Gabriel Cramer, a Swiss mathematician born in Geneva on 31 July 1704, published the rule in 1750.
- For \( a_1x + b_1y + c_1z = d_1 \) and so on, \( \Delta \) is the determinant of the coefficients; \( \Delta_x, \Delta_y, \Delta_z \) replace the \( x, y, z \) column by the constants.
- \( x = \frac{\Delta_x}{\Delta}, y = \frac{\Delta_y}{\Delta}, z = \frac{\Delta_z}{\Delta} \), provided \( \Delta \neq 0 \).
- Advantage: any one unknown can be found without the others.
- Book applications: prices of books, costs of wheat, sugar and rice, rates of commission, investments at different interest rates.
1.3 Transition probability matrices
- A one-stage transition probability is the probability of moving from state \( S_n \) to \( S_{n+1} \) in one unit of time.
- Entries satisfy \( P_{jk} \geq 0 \) and each row adds up to 1.
- Share after one period = (present share row) \( \times T \); after two periods multiply by \( T \) again.
- Equilibrium: \( (A \; B)T = (A \; B) \) with \( A + B = 1 \).
Worked example: for \( T = \begin{pmatrix} 0.9 & 0.1 \\ 0.3 & 0.7 \end{pmatrix} \), \( 0.9A + 0.3(1 - A) = A \) gives \( 0.4A = 0.3 \), so \( A = \frac{3}{4} \): brand A holds 75% and brand B 25% at equilibrium.
Exam tips
- Always state the upper bound \( \rho(A) \leq \min\{m, n\} \) before reducing.
- For "find k" problems, reduce \( [A, B] \) to echelon form and make the last row consistent.
- Check that each row of a transition matrix adds up to 1 before using it.