In statistics, the Bhattacharyya distance measures the similarity of two probability distributions. If two probability distributions are same, then in this case what will be coefficient value.
There are lots of these called goodness of fit tests e.g. look at the Kolmogorov-Smirnov test which does the same kind of thing. Best, David Booth
Actually, It depends on you data and your distribution. You should make a goodness-of-fit test based on Bhattacharya distance then by calculating critical values of the the test via Monte-Carlo simulation you can infer about similarity of two probability distributions.
It is between 0 and 1. 0 for the cases that there is no similarity (lack of common sample space), and 1 for when the two distributions/densities are exactly the same.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python