Existence of Inverse Elementary Column Operation

From ProofWiki
Jump to navigation Jump to search

Theorem

Let $\map \MM {m, n}$ be a metric space of order $m \times n$ over a field $K$.

Let $\mathbf A \in \map \MM {m, n}$ be a matrix.

Let $\map e {\mathbf A}$ be an elementary column operation which transforms $\mathbf A$ to a new matrix $\mathbf A' \in \map \MM {m, n}$.

Let $\map {e'} {\mathbf A'}$ be the inverse of $e$.


Then $e'$ is an elementary column operation which always exists and is unique.


Proof

Let us take each type of elementary column operation in turn.


For each $\map e {\mathbf A}$, we will construct $\map {e'} {\mathbf A'}$ which will transform $\mathbf A'$ into a new matrix $\mathbf A \in \map \MM {m, n}$, which will then be demonstrated to equal $\mathbf A$.


In the below, let:

$\kappa_k$ denote column $k$ of $\mathbf A$
$\kappa'_k$ denote column $k$ of $\mathbf A'$
$\kappa_k$ denote column $k$ of $\mathbf A$

for arbitrary $k$ such that $1 \le k \le m$.


By definition of elementary column operation:

only the column or columns directly operated on by $e$ is or are different between $\mathbf A$ and $\mathbf A'$

and similarly:

only the column or columns directly operated on by $e'$ is or are different between $\mathbf A'$ and $\mathbf A$.

Hence it is understood that in the following, only those columns directly affected will be under consideration when showing that $\mathbf A = \mathbf A$.


$\text {ECO} 1$: Scalar Product of Column

Let $\map e {\mathbf A}$ be the elementary column operation:

$e := \kappa_k \to \lambda \kappa_k$

where $\lambda \ne 0$.

Then $\kappa'_k$ is such that:

$\forall a'_{k i} \in \kappa'_k: a'_{k i} = \lambda a_{k i}$


Now let $\map {e'} {\mathbf A'}$ be the elementary column operation which transforms $\mathbf A'$ to $\mathbf A$:

$e' := \kappa_k \to \dfrac 1 \lambda \kappa_k$

Because it is stipulated in the definition of an elementary column operation that $\lambda \ne 0$, it follows by definition of a field that $\dfrac 1 \lambda$ exists.

Hence $e'$ is defined.

So applying $e'$ to $\mathbf A'$ we get:

\(\ds \forall a_{i k} \in \kappa_k: \, \) \(\ds a_{i k}\) \(=\) \(\ds \dfrac 1 \lambda a'_{i k}\)
\(\ds \) \(=\) \(\ds \dfrac 1 \lambda \paren {\lambda a_{i k} }\)
\(\ds \) \(=\) \(\ds a_{i k}\)
\(\ds \leadsto \ \ \) \(\ds \forall a_{i k} \in \kappa_k: \, \) \(\ds a_{i k}\) \(=\) \(\ds a_{i k}\)
\(\ds \leadsto \ \ \) \(\ds \kappa_k\) \(=\) \(\ds \kappa_k\)
\(\ds \leadsto \ \ \) \(\ds \mathbf A\) \(=\) \(\ds \mathbf A\)

It is noted that for $e'$ to be an elementary column operation, the only possibility is for it to be as defined.

$\Box$


$\text {ECO} 2$: Add Scalar Product of Column to Another

Let $\map e {\mathbf A}$ be the elementary column operation:

$e := \kappa_k \to \kappa_k + \lambda r_l$

Then $\kappa'_k$ is such that:

$\forall a'_{i k} \in \kappa'_k: a'_{i k} = a_{i k} + \lambda a_{i l}$


Now let $\map {e'} {\mathbf A'}$ be the elementary column operation which transforms $\mathbf A'$ to $\mathbf A$:

$e' := \kappa'_k \to \kappa'_k - \lambda \kappa'_l$


Applying $e'$ to $\mathbf A'$ we get:

\(\ds \forall a_{i k} \in \kappa_k: \, \) \(\ds a_{i k}\) \(=\) \(\ds a'_{i k} - \lambda a'_{i l}\)
\(\ds \) \(=\) \(\ds \paren {a_{i k} + \lambda a_{i l} } - \lambda a'_{i l}\)
\(\ds \) \(=\) \(\ds \paren {a_{i k} + \lambda a_{i l} } - \lambda a_{i l}\) as $\lambda a'_{i l} = \lambda a_{i l}$: column $l$ was not changed by $e$
\(\ds \) \(=\) \(\ds a_{i k}\)
\(\ds \leadsto \ \ \) \(\ds \forall a_{i k} \in \kappa_k: \, \) \(\ds a_{i k}\) \(=\) \(\ds a_{i k}\)
\(\ds \leadsto \ \ \) \(\ds \kappa_{i k}\) \(=\) \(\ds \kappa_{i k}\)
\(\ds \leadsto \ \ \) \(\ds \mathbf A\) \(=\) \(\ds \mathbf A\)


It is noted that for $e'$ to be an elementary column operation, the only possibility is for it to be as defined.

$\Box$


$\text {ECO} 3$: Exchange Columns

Let $\map e {\mathbf A}$ be the elementary column operation:

$e := \kappa_k \leftrightarrow \kappa_l$

Thus we have:

\(\ds \kappa'_k\) \(=\) \(\ds \kappa_l\)
\(\, \ds \text {and} \, \) \(\ds \kappa'_l\) \(=\) \(\ds \kappa_k\)


Now let $\map {e'} {\mathbf A'}$ be the elementary column operation which transforms $\mathbf A'$ to $\mathbf A$:

$e' := \kappa'_k \leftrightarrow \kappa'_l$


Applying $e'$ to $\mathbf A'$ we get:

\(\ds \kappa_k\) \(=\) \(\ds \kappa'_l\)
\(\, \ds \text {and} \, \) \(\ds \kappa_l\) \(=\) \(\ds \kappa'_k\)
\(\ds \leadsto \ \ \) \(\ds \kappa_k\) \(=\) \(\ds \kappa_k\)
\(\, \ds \text {and} \, \) \(\ds \kappa_l\) \(=\) \(\ds \kappa_l\)
\(\ds \leadsto \ \ \) \(\ds \mathbf A\) \(=\) \(\ds \mathbf A\)

It is noted that for $e'$ to be an elementary column operation, the only possibility is for it to be as defined.

$\Box$


Thus in all cases, for each elementary column operation which transforms $\mathbf A$ to $\mathbf A'$, we have constructed the only possible elementary column operation which transforms $\mathbf A'$ to $\mathbf A$.

Hence the result.

$\blacksquare$


Also see