search

weecology / bbs-forecasting

Research on forecasting using Breeding Bird Survey data
MIT License
7 stars 9 forks source link

New Fig: Forecast Skill & Richness Change #256

Open ethanwhite opened 6 years ago

ethanwhite commented 6 years ago

screenshot from 2018-03-06 10-04-52

davharris commented 6 years ago

We should think about what kinds of inference are valid here. The key phrase in the stats literature is “selecting on a dependent variable”. I’m not sure if that’s a problem here, but we should look into it.

Dave

On Mar 6, 2018, at 10:23 AM, Ethan White notifications@github.com wrote:

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

ethanwhite commented 6 years ago

We may want to make this figure (or another version) include sign as well as magnitude since it could matter if there is a difference in our ability to forecast increases vs decreases in richness.

2018-03-07 09 43 32