In [[hypothesis testing]], a simple hypothesis is one that completely specifies the distribution. A hypothesis that is not simple is called a composite hypothesis. While it is not strictly correct, we will use simple hypothesis to mean a hypothesis of the form $H = \theta$ and composite hypothesis to mean any of $\begin{align}H \ne \theta && \text{or} && H > \theta && \text{or} && H < \theta\end{align}$ Strictly speaking, a hypothesis of the form $H_0 = \theta_0$ is not simple if the distribution is specified by multiple parameters (e.g., mean and variance), one or more of which may remain unknown after the hypothesis test is conducted.