search

numbats / wages-refresh

This will be a repo for re-building, and documenting the process of the wages data from the NLSY
MIT License
0 stars 0 forks source link

bam instead of gam #2

Closed Dewi-Amaliah closed 3 years ago

Dewi-Amaliah commented 3 years ago

Is it okay to use bam instead of gam to model the data? I have used gam but it resulted in this error:

"vector memory exhausted (limit reached?)"

However, when trying bam, it has been running for 4 hours and has not completed

dicook commented 3 years ago

Yes, or regression trees would be ok too. Or try using a GLM, with a subject random effect.

You can also try to fit a GAM on a small subset to see if it will work.

Actually always a good idea to fit a model to a small subset as a start.

Sent from my iPhone

On 21 Dec 2020, at 5:34 pm, Dewi Amaliah notifications@github.com wrote:

 Is it okay to use bam instead of gam to model the data? I have used gam but it resulted in this error:

"vector memory exhausted (limit reached?)"

However, when trying bam, it has been running for 4 hours and has not completed

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