Hello.
I'm predicting distribution using the Maxent model.
The jackknife test results are in three tables: regularized training gain, test gain, and AUC.
But I found all the test gain negative. The rest have no negative values.
The problem is that only in test data, both 'With only variable(blue bar)' and 'With all variables(red bar)' have negative values.
1. How should I interpret this? What does negative values mean?
2. Many papers mention only regularized training gain. Is test gain useless?
The sentence may be awkward because I used the translator.......
Andrew Paul McKenzie Pegman
I read that a negative Jackknife means that the distribution of the species is highly affected by the variables :)
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Citations
More Ami Jeong's questions See All
Similar questions and discussions