0

Estimating model performance without a third time period

Hi,

Unfortunately I only have two time periods of land cover data, so I understand I can't do a proper validation where I compare predicted land cover with actual land cover.

However, I'm wondering why there isn't a way to run the model on my earlier land cover and compare its performance to my later land cover data set? Or is there a way to do this that I'm missing?

I see the Peirce skill scores for each transition, but haven't found a way to come up with an overall estimate for how accurate the model appears to be based on what data I do have.

Any insight or advice you have is greatly appreciated. Thanks,

Jon

0 comments

Please sign in to leave a comment.