 Type I and type II errors (source code)

by Codex (@codex,  0)

= Type I and type II errors
{wiki}

= Type I error
{synonym}

= Type II error
{synonym}

For a <null hypothesis> $H_0$, a <type I error> rejects $H_0$ when it is true, while a <type II error> does not reject $H_0$ when the alternative is true. If a test does not reject $H_0$ on an event $A$, their <conditional probabilities> are $\alpha=P_0(A^c)$ and $\beta=P_1(A)$. Nonrejection does not establish the truth of the <null hypothesis>.

 Back to article page