Direct answer
Matrices organise linear relationships; valid operations depend on their dimensions, and multiplication composes transformations in an order that generally cannot be reversed.
What this calculation tells you
A matrix can represent a linear transformation, a system of coefficients, or structured data. Addition combines same-sized matrices entry by entry, while multiplication forms row-by-column combinations.
For a square matrix, the determinant indicates scaling and singularity information. A zero determinant means no ordinary inverse exists.
Where it is used
Geometry and graphics
Compose linear scaling, rotation, and shear transformations.
Equation systems
Represent coefficients and solve when the system and method permit.
Science and engineering
Organise linear models with units and assumptions attached to rows and columns.
Data analysis
Express transformations and covariance-like structures without confusing them with raw meaning.
When this guide helps
- Checking whether two matrices can multiply.
- Composing transformations in the correct order.
- Testing whether a square matrix is singular.
- Interpreting a computed inverse within a model.
Start with dimensions
An m-by-n matrix can multiply an n-by-p matrix, producing an m-by-p result. The inner dimensions match; the outer dimensions describe the result.
Treat multiplication as composition
Changing the order usually changes both validity and meaning. For column-vector conventions, the rightmost transformation acts first.
Interpret determinant and inverse carefully
A determinant near zero can also signal numerical sensitivity even when it is not exactly zero. An inverse is a mathematical object, not automatic evidence that a real model is appropriate.
Common mistakes
Common errors include multiplying entries position by position and assuming AB equals BA.
- Write dimensions beside each matrix.
- Keep row and column meanings consistent.
- Check singularity and numerical scale.
Worked case: order-sensitive multiplication
Let A=[[1, 2], [3, 4]] and B=[[2, 0], [1, 2]].
Row-by-column products give AB=[[1x2+2x1, 1x0+2x2], [3x2+4x1, 3x0+4x2]]=[[4, 4], [10, 8]].
AB=[[4, 4], [10, 8]].
Every entry has an auditable row-column dot product.
Reproduce this worked caseOpen Matrix Calculator
Worked case: reverse the order
Multiply the same matrices as BA.
BA=[[2x1+0x3, 2x2+0x4], [1x1+2x3, 1x2+2x4]]=[[2, 4], [7, 10]].
BA differs from AB.
Matrix multiplication is generally not commutative even when both products are defined.
Reproduce this worked caseOpen Matrix Calculator
Compare matrix operations cases before generalising
Dimensions determine whether an operation is defined, while labels determine whether it is meaningful. An inverse exists only for a square nonsingular matrix and should be verified with an identity product.
| Operation | Result |
|---|---|
| AB | [[4, 4], [10, 8]] |
| BA | [[2, 4], [7, 10]] |
| det(A) | 1x4-2x3=-2 |
matrix operations: calculation checklist
- Write dimensions
- Preserve multiplication order
- Use row-by-column products
- Check determinant/invertibility when relevant
- Verify with identity or independent multiplication
Practical questions
Frequently asked questions
Can any two matrices be added?
No. Matrix addition requires the same dimensions so corresponding entries align.
Why can multiplication order matter?
Each product composes row and column relationships in a different sequence, and the reversed product may have another size or meaning.
Does every square matrix have an inverse?
No. A square matrix is invertible only when it is nonsingular, equivalently when its determinant is nonzero in exact arithmetic.
Further reading
Authoritative sources
Use these primary and professional resources to check definitions, conventions, or requirements that may extend beyond this guide.