MantisBT - MEGA
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0006787MEGA[All Projects] Feedbackpublic2025-05-09 13:442025-05-09 13:44
Reporterguest 
Assigned To 
PrioritynormalSeverityminorReproducibilityhave not tried
StatusnewResolutionopen 
PlatformOSOS Version
Product VersionMEGA 12 
Target VersionFixed in Version 
First NameHolger
Last NameHerlyn
Emailherlyn@uni-mainz.de
Confirm Emailherlyn@uni-mainz.de
Summary0006787: AICc inferecnce in model testing
DescriptionDear colleagues,
In model testing I tried to re-caculate the AICc values presented by MEGA 12. For doing so, I fed k and lnL accordiing to the output table into the standard formula: AICc = AIC + (2k(k + 1)) / (n - k - 1), where AIC = 2k - 2lnL and n = number of nucleotide positions of the alignment. The resulting AICc values differ from the ones reported by MEGA 12. Is this perhaps because MEGA uses a different formula for AICc and/or AIC inferencce and, if so which? THANK YOU!!!
Cheers, Holger
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2025-05-09 13:44guestNew Issue

There are no notes attached to this issue.