Determinants — cofactor expansion along any row - column

3. Derivation from first principles (WHY it works)

HOW we get row-1 expansion. Group the $n!$ permutations by which column the row-1 entry lands in, call it $j=\sigma(1)$ .

$\det A=\sum_{j=1}^{n} a_{1j}\!\!\sum_{\substack{\sigma\in S_n\\ \sigma(1)=j}}\!\!\operatorname{sgn}(\sigma)\prod_{k=2}^{n} a_{k,\sigma(k)}.$

Why this step? We just split one giant sum into $n$ buckets — no terms lost, none added.

Now look at the inner sum. With column $j$ and row $1$ removed, the remaining $\sigma$ restricted to rows $2..n$ ranges over all permutations of the leftover columns — that's exactly the definition of the minor $M_{1j}$ , except for a sign.

The sign: to bring entry $(1,j)$ into the leading position you must hop the chosen column past $j-1$ columns; combined with row $1$ this contributes a factor $(-1)^{1+j}$ . Hence: $\sum_{\substack{\sigma(1)=j}}\operatorname{sgn}(\sigma)\prod_{k\ge2}a_{k,\sigma(k)}=(-1)^{1+j}M_{1j}=C_{1j}.$

Why this step? The parity bookkeeping is exactly what $(-1)^{i+j}$ packages. Substituting back: $\boxed{\det A=\sum_{j=1}^{n}a_{1j}C_{1j}.}$

Any other row/column: the Leibniz sum is symmetric — every row and column appears identically. Swapping two rows flips $\operatorname{sgn}$ but you can group by any row first. Since $\det A=\det A^{T}$ (transpose leaves the permutation sum invariant), column expansion = row expansion of the transpose. Therefore every line gives the same answer. ∎

Figure — Determinants — cofactor expansion along any row - column

4. The 80/20 — what actually wins marks

5. Worked examples

6. Steel-manned mistakes

7. Flashcards

What is the minor

M_{ij}

\det

of the matrix after deleting row

i

and column

j

What is the cofactor

C_{ij}

(-1)^{i+j}M_{ij}

— the signed minor.

State cofactor expansion along row

i

\det A=\sum_{j} a_{ij}(-1)^{i+j}M_{ij}

Can you expand along any row or column?

Yes — all give the same value (follows from Leibniz sum &

\det A=\det A^T

Sign of cofactor

C_{12}

(-1)^{1+2}=-1

2\times2

determinant formula?

ad-bc

Best line to expand along (80/20)?

The row/column with the most zeros.

Effect of swapping two rows on

\det

Multiplies determinant by

-1

Effect of adding

k\times

one row to another?

No change to

\det

Why does

(-1)^{i+j}

appear?

It is the parity (sign) of the permutation that moves entry

(i,j)

to the leading corner.

What does

\det A=0

mean geometrically?

The map collapses space →

A

is singular/non-invertible.

Recall Feynman: explain to a 12-year-old

Imagine a big grid of numbers. A "determinant" is one special number that tells you how much a shape grows or shrinks when you stretch it with this grid. To find it, you walk along the top row. For each number, you cover up its row and column with your hand, find the little number left over, give it a $+$ or $-$ sign like a checkerboard, and add everything. Cool part: you can walk along any row or column — even one full of zeros (which is easiest, because zeros mean less work!) — and you get the same answer every time.

Connections

Leibniz formula for determinants — the parent definition we derived from.
Determinant via row reduction — the faster computational cousin.
Properties of determinants (multilinearity, alternation)
Adjugate matrix and inverse — $A^{-1}=\frac{1}{\det A}\operatorname{adj}(A)$ , built from cofactors.
Cramer's rule — solves $Ax=b$ using cofactors.
Invertibility and singular matrices — $\det A\ne0 \iff$ invertible.

Concept Map

Hinglish (regional understanding)

Intuition Hinglish mein samjho

Dekho, determinant ek square matrix se nikla hua ek single number hota hai jo batata hai ki matrix space ko kitna stretch ya squeeze karti hai. Agar yeh number $0$ ho gaya, matlab matrix ne space ko "flatten" kar diya — woh invertible nahi hai. Cofactor expansion ek recursive trick hai: bade $n\times n$ determinant ko chhote $(n-1)\times(n-1)$ pieces mein tod do.

Recipe simple hai: kisi bhi ek row ya column ko pakdo. Uske har entry ke liye, us entry ki row aur column ko dhak do (hide karo), jo chhota determinant bacha woh minor $M_{ij}$ hai. Phir checkerboard sign $(-1)^{i+j}$ laga do — yeh cofactor $C_{ij}$ ban gaya. Sab entries ko apne cofactor se multiply karke add kar do — bas, determinant mil gaya!

Sabse important baat (aur exam ka 80/20): tum kisi bhi row ya column se expand kar sakte ho, answer hamesha same aata hai. Isliye smart move yeh hai — jis line mein sabse zyada zeros hon usi se expand karo, kyunki zero wala term toh khud hi gayab ho jaata hai, kaam kam! Aur agar zeros naa hon, toh pehle ek row ko doosri mein add/subtract karke (jo determinant change nahi karta) zeros bana lo, phir expand karo.

Common galti: minor aur cofactor ko same samajh lena. Sign $(-1)^{i+j}$ lagana mat bhoolo. Aur yaad rakho — row swap karne se determinant ka sign ulta ho jaata hai, par row add karne se kuch nahi badalta. Yeh teen rules (swap = $-1$ , scale = $k$ , add = no change) pakke kar lo.

Determinants — cofactor expansion along any row - column

1. Building blocks

2. The expansion formula

3. Derivation from first principles (WHY it works)

4. The 80/20 — what actually wins marks

5. Worked examples

6. Steel-manned mistakes

7. Flashcards

Connections

Concept Map

Hinglish (regional understanding)

Go deeper — visual, from zero

Test yourself — Linear Algebra (Full)

Connections