| Author: | Enys Mones |
|---|---|
| Version: | 0.1 |
| License: | MIT |
This python script (model.py) implements model selection for discrete distributions. As being motivated by degree distributions in complex networks, currently it supports only distributions over non-negative integers and the following models:
- Poisson
- exponential
- log-normal
- Weibull
- shifted power-law
- truncated power-law (power-law with cutoff)
- normal
It requires the following python packages:
csvargparsempmathscipynumpy
Passing --help will print the help menu.
One-column CSV, with the numbers being the single sample values from the distribution.
The script is shipped with some basic testing methods which can be accessed by the corresponding commands when started.
Distribution of the original data and the optimal theoretical distribution of all models.
More distributions...
| [1] | Aaron Causet, Cosma Rohilla Shalizi and M. E. J. Newman: Power-law distributions in empirical data. SIAM Review 51 (4) (2009): 661-703. |
| [2] | M. P. H. Stumpf and P. J. Ingram: Probability models for degree distributions of protein interaction networks. Europhys. Lett. 71 (1) (2005): 152-158. |
| [3] | Gideon Schwarz: Estimating the dimension of a model. The Annals of Statistics 6 (2) (1978): 461-464. |