Joachim Domsta
Dear Ilya,
Looks simple, but strange.
Simple since this is (as a function of variable x) a convex combination of the truncated normal N(0; \lamba_+) to [0,\infty) and N(0; \lambda_-) to (-\infty, 0], with densities
2 \lambda exp(- \lambda^2 x^2/2)/\sqrt{2 \pi}, each correspongly with \lambda=\lambda_+ and \lambda_-
Strange, since their two coefficients \sqrt{2\pi} \lambda^{\alpha} / 2^{(\alpha - ??)/2} \Gamma(, , ,) do not sum up to one.
Perhaps this should be a two-dimensional pd?
Regards, Joachim
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R
Similar questions and discussions