BAYES_BETA
Bayesian Parameter Estimation


BAYES_BETA if a FORTRAN90 program which is a simple Bayesian Statistics demonstration.

Suppose we're watching a "system" and trying to analyze its behavior. Each time we observe the system, it flips a coin a certain number of times, and reports the number of heads and tails. We want to estimate THETA1 and THETA2, the probabilities of heads and of tails.

We treat the values of THETA1 and THETA2 as random variables themselves, controlled by a Beta probability density function, which has parameters ALPHA1 and ALPHA2. We make an arbitrary or informed guess for initial values of ALPHA1 and ALPHA2. We observe the system, and adjust ALPHA1 and ALPHA2 using Bayes's Law. We continue until we are satisfied that our estimates seem to have converged.

Licensing:

The computer code and data files made available on this web page are distributed under the GNU LGPL license.

Languages:

BAYES_BETA is available in a FORTRAN90 version.

Related Data and Programs:

BAYES_DICE, a FORTRAN90 program which uses Bayesian analysis to adjust a model of loaded dice based on a sequence of experimental observations.

BAYES_WEIGHT, a FORTRAN90 program which uses Bayesian analysis to adjust a model of loaded dice based on a sequence of experimental observations.

BDMLIB, a FORTRAN90 library which estimates the weights in a Dirichlet mixtured based on sample data;

Source Code:

Examples and Tests:

List of Routines:

You can go up one level to the FORTRAN90 source codes.


Last revised on 30 August 2005.