inverse gamma prior More...
#include <prior.h>
Inheritance diagram for invGammaPrior:
Public Member Functions | |
| invGammaPrior (double shape, double scale) | |
| Initialize inverse gamma prior. | |
| invGammaPrior (const invGammaPrior &original) | |
| copy constructor | |
| virtual double | pdf (double x) |
| virtual double | dpdf (double x) |
| evaluate the derivative of the pdf of the prior at position x (in this default form, the parameter is completely unconstrained) | |
| virtual double | rand (void) |
| draw a random number | |
| PsiPrior * | clone (void) const |
| clone by value | |
| virtual double | mean (void) const |
| return the mean | |
| double | std (void) const |
| return the standard deviation | |
| virtual void | shrink (double xmin, double xmax) |
| shrink the prior if it is broader than the range between xmin and xmax | |
| virtual int | get_code (void) const |
| Doesn't shrink!! | |
Detailed Description
inverse gamma prior
This defines an inverse gamma prior that is conjugate to the variance of a normal distribution and can probably be assumed quasi conjugate to the width or parameters like that. Its pdf is defined by
![\[ f(x) = \frac{\beta^\alpha}{\Gamma(\alpha)} x^{-\alpha-1} \exp ( -\beta/x ) \]](form_7.png)
The documentation for this class was generated from the following file:
- src/prior.h
Generated on Mon Jul 4 2011 14:52:04 for Psi++ by
1.7.1