Mode for posterior

[amroco]

Hi there, I know it is not ideal to use a single point statistic to summarise a posterior distribution. I am deriving the estimated contributions of 4 sources (primary production pathways) to a variety of marine taxa. To get an overall estimate of the contribution of each pathway to the community in general I will then use survey abundance data to which I apply a rough estimate of pathway use according to the mixing model. For instance I may show that all fish species in the community use mostly pelagic sources, therefore when looking at the community level (regardless of species) the pelagic pathway will contribute to a majority of overall community. I know this method has holes, but as a baseline it is useful.

The Summary Statistics from my MixSIAR model give me the Mean, SD and variaous confidence intervals. However, given that many of my posterior distributions are skewed, I would rather use the mode.

Is there a way for me to get those values from the summary statistic?

Thank you for your help.

[brianstock]

See replies in other issues and this vignette: http://brianstock.github.io/MixSIAR/articles/modify_output.html

<>$BUGSoutput$sims.list$<> has the mcmc draws so you can calculate whatever summary stat you want

On Wed, Jun 7, 2023, 4:03 AM Amy Rose Coghlan @.***> wrote:

Hi there, I know it is not ideal to use a single point statistic to summarise a posterior distribution. I am deriving the estimated contributions of 4 sources (primary production pathways) to a variety of marine taxa. To get an overall estimate of the contribution of each pathway to the community in general I will then use survey abundance data to which I apply a rough estimate of pathway use according to the mixing model. For instance I may show that all fish species in the community use mostly pelagic sources, therefore when looking at the community level (regardless of species) the pelagic pathway will contribute to a majority of overall community. I know this method has holes, but as a baseline it is useful.

The Summary Statistics from my MixSIAR model give me the Mean, SD and variaous confidence intervals. However, given that many of my posterior distributions are skewed, I would rather use the median or mode.

Is there a way for me to get those values from the summary statistic?

Thank you for your help.

— Reply to this email directly, view it on GitHub https://github.com/brianstock/MixSIAR/issues/347, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABHDA423PV4PWDJ6Z5LCNCTXJ7OQJANCNFSM6AAAAAAY5FZNHM . You are receiving this because you are subscribed to this thread.Message ID: @.***>