Closed yuqing19118 closed 8 years ago
Thanks. I would like to add you to the contributor list of phylolm. If you agree, please let me know your full name.
Best, Lam
Hello Lam,
Thanks so much for your prompt response. My name is Qing Yu (Sabrina)
Thanks
Sent from my iPhone
On Jun 13, 2016, at 2:44 PM, Lam Ho notifications@github.com<mailto:notifications@github.com> wrote:
Thanks. I would like to add you to the contributor list of phylolm. If you agree, please let me know your full name.
Best, Lam
You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/lamho86/phylolm/pull/8#issuecomment-225686915, or mute the threadhttps://github.com/notifications/unsubscribe/ASf_3Qq4T-0UKB09a_9NTnzrYflkvHqeks5qLbMvgaJpZM4I0eIb.
Also, I added the description of model for measurement error in the manual. I disabled lambda plus measurement error, because there are identifiability problems. If you want to see for yourself, Below is an example. You would need to re-enable lambda + measurement error to run this example (comment out lines 12 and 13 in phylolm.R). The example shows that the likelihood profile is flat where it's maximum.
first: simulate some data
next: use phylolm with a fixed lambda, to get the profile likelihood at many fixed lambda values
the plot above showed the flat likelihood. The plot below shows how lambda hat varies with sigma_err hat: