I have constructed a GAM which predicts the probability of presence-absence data as a continuous probability where zero=low probability of species presence and one=high probability of species presence.
I have seen the use of the AUC as a validation technique for binary classification models, however in all my searching have yet to find a statistical measure which can help me measure the predictive accuracy of my model.
Are there any statistical tests anyone knows of that I could use to validate a model which is predicting the probability of a presence-absence spatial dataset?
A. U. Usman
You can look into this:
https://www.hackerearth.com/practice/machine-learning/machine-learning-algorithms/logistic-regression-analysis-r/tutorial/
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Condensed Matter Physics
- Zero
Similar questions and discussions