= LU decomposition of an upper Hessenberg matrix
{c}
{title2=$A=LU$}
When <Gaussian elimination> without <row interchanges> is defined, an <upper Hessenberg matrix> needs at most one elimination per <pivot column>, because only its first subdiagonal can be nonzero. The <LU decomposition> has a unit lower <bidiagonal matrix> $L$ and an <upper triangular matrix> $U$. If the input <matrix> is also symmetric, it is a <tridiagonal matrix> and $U$ is upper bidiagonal.
The nonzero-<pivot> condition is essential: $\left(\begin{smallmatrix}0&1\\1&0\end{smallmatrix}\right)$ is upper Hessenberg and invertible but has no such unpivoted <LU decomposition>. A pivoted factorization can still be computed, with its permutations recorded explicitly.
Back to article page