Back-transformation of phylogenetic logistic regression

[rosbakh]

Dear developers of the phylolm package,

I am sorry for posting this question here - I could not find any help elsewhere. For the sake of clarity, I would like to back-transform the estimates for a phylogenetic logistic regression I run with the help of your package. When I apply the 'standard' exp () I used to use for non-phylogenetic regressions, I get meaningless results. Am I missing something? Should I back-transform the estimates differently?

Thanks, Sergey

[lamho86]

Hello Sergey,

I am still unsure about your question. Could you please give more details?

Lam

Lam Ho | Assistant Professor FACULTY OF SCIENCE Department of Mathematics and Statistics 902.494.1069

Dalhousie University dal.ca

On Jan 23, 2020, at 3:46 PM, rosbakh notifications@github.com wrote:

Dear developers of the phylolm package,

I am sorry for posting this question here - I could not find any help elsewhere. For the sake of clarity, I would like to back-transform the estimates for a phylogenetic logistic regression I run with the help of your package. When I apply the 'standard' exp () I used to use for non-phylogenetic regressions, I get meaningless results. Am I missing something? Should I back-transform the estimates differently?

Thanks, Sergey

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/lamho86/phylolm/issues/29?email_source=notifications&email_token=ABW3HP3DGCF3ZW5SWZGYND3Q7HXYPA5CNFSM4KK37BX2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IIK66JQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABW3HP2QSNSN473LBRCGRB3Q7HXYPANCNFSM4KK37BXQ.

[rosbakh]

Hi Liam,

Sorry, I made some mistakes in the code that resulted in wrong estimates. It works well now.

Take care, Sergey

[rosbakh]

Hi Liam,

Sorry, I made some mistakes in the code that resulted in wrong estimates. It works well now.

Take care, Sergey