= Finite-sample maximum bias
{title2=$b_n(m;y)=\sup_{z:d_H(z,y)\leq m}|T_n(z)-T_n(y)|$}
Here $d_H$ counts the observations replaced. This curve measures the largest change after at most $m$ arbitrary replacements. Under the first-unbounded convention, the <replacement breakdown point> is $n^{-1}\min\{m:b_n(m;y)=\infty\}$. The largest fraction still guaranteed bounded is one grid step smaller; those conventions must not be confused. A <sample mean> breaks after one replacement, while a scalar <sample median> with a fixed ordinary tie convention needs about half the sample replaced.
Back to article page
