Listy Biometryczne - Biometrical Letters Vol. 32(1995), No. 1, 15-21
|
BAYESIAN ESTIMATION OF A STRUCTURAL CHANGES MODEL WITH DEPENDENT BERNOULLI VARIABLE Krystyna Pruska Chair of Statistical Methods, University of Lodz, 41, Rewolucji 1905r., 90-214-Lodz |
In this paper a structural changes model is considered. Two Bernoulli variables are in the model. These variables occur with the probability X and 1-\ respectively. The Pl and/>2 are distribution parameters of the Bernoulli variables.
The Bayesian estimators of this model parameters are derived in two cases:
1) when X. is known and pj, p2 are unknown,
2) when K is unknown andp1} p% are known,
The prior distributions of X, pi,p% are uniform distributions on the interval (0,1).
The Bayesian estimators of this model parameters are derived in two cases:
1) when X. is known and pj, p2 are unknown,
2) when K is unknown andp1} p% are known,
The prior distributions of X, pi,p% are uniform distributions on the interval (0,1).
