The half-Cauchy is defined on and is Cauchy distributed on this domain. It can also be interpreted as a mixture of inverse gamma distributions[]]. In other words:
\begin{equation*}
X \sim \operatorname{C}^{+}(0, a)
\end{equation*}
is equivalent to
\begin{align*}
X^{2} &\sim \operatorname{InverseGamma}(1/2,\, 1/\xi)\\
\xi &\sim \operatorname{InverseGamme}(1/2,\, 1/a^2)
\end{align*}
References
- High-Dimensional Bayesian Regularised Regression with the BayesReg Package (ArXiv)