search

ocbe-uio / TruncExpFam

R package to generate data related to the Truncated Exponential Family
https://ocbe-uio.github.io/TruncExpFam/
GNU General Public License v3.0
0 stars 1 forks source link

Fix support check for Inverse Gaussian #106

Closed wleoncio closed 8 months ago

wleoncio commented 8 months ago

This should fail, since the support for the Inverse Gaussian is $(0, \infty)$.

validateSupport.trunc_invgauss(n = 0, NULL)

It does indeed fail on rmutil::prinvgauss()

rmutil::pinvgauss(q = 0, m = 1, s = 2)

The former does not fail because judgeSupportLimits() doesn't distinguish between $< 0$ and $\le 0$ in this particular case.

rho62 commented 8 months ago

I think it’s best that it actually does fail, in the same way that we expect an error message if issue log(0) /René

Fra: Waldir Leoncio @.> Svar til: ocbe-uio/TruncExpFam @.> Dato: fredag 1. mars 2024 kl. 17:34 Til: ocbe-uio/TruncExpFam @.> Kopi: Subscribed @.> Emne: [ocbe-uio/TruncExpFam] Fix support check for Inverse Gaussian (Issue #106)

This should fail, since the support for the Inverse Gaussianhttps://en.wikipedia.org/wiki/Inverse_Gaussian_distribution is $(0, \Inf)$.

validateSupport.trunc_invgauss(n = 0, NULL)

It does indeed fail on rmutil::prinvgauss()

rmutil::pinvgauss(q = 0, m = 1, s = 2)

The former does not fail because judgeSupportLimits() doesn't distinguish between $< 0$ and $\le 0$ in this particular case.

— Reply to this email directly, view it on GitHubhttps://github.com/ocbe-uio/TruncExpFam/issues/106, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AFPRUPXERI23WXEIZBHBFDDYWCUYZAVCNFSM6AAAAABECBTXDSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGE3DGNZVG42TONI. You are receiving this because you are subscribed to this thread.Message ID: @.***>