search

katcatalano / gordita

body size and biomass distribution fitting
0 stars 0 forks source link

truncated distribution #1

Open katcatalano opened 5 years ago

katcatalano commented 5 years ago

alternative method of fitting distributions from Ed:

"You can also use a truncated distribution, setting min and max as some quantiles and then setting the probabilities for max to be zero. You then have to normalize it by dividing by the area remaining (eg. 0.95 or 0.99). Finally you multiply the entire distribution by the total taxon biomass to get the biomass density function."

this link seems to show how to do this:

katcatalano commented 5 years ago

Talked to Ed 9/6 (notes in star notebook) used the mean equation rather the mode, because that's what the values Ed had in the spread sheet were. Doing this decreased the spread for plants a tiny bit... then made a weird looking distribution for bacteria. Stopped there. Probably need to frame this more as a maximum likelihood problem?