As the log transformation is monotone, the density function can be derived with a simple Jacobian transformation. Taking y = log x, dy = dx/x = e-ydx, the transformed inverse gamma density is fY(y) = fX(ey) e–y = [θk / Γ(k)] exp(–k y – θ e-y).
As the log transformation is monotone, the density function can be derived with a simple Jacobian transformation. Taking y = log x, dy = dx/x = e-ydx, the transformed inverse gamma density is fY(y) = fX(ey) e–y = [θk / Γ(k)] exp(–k y – θ e-y).
The inverse gamma distribution is a two-parameter family of continuous probability distributions on the positive real line, which is the distribution of the reciprocal of a variable distributed according to the gamma distribution.