摘要：This paper proposes the construction of a Bayesian speci cation test
based on the encompassing principle for the case of partial observability of latent
variables. A structural parametric model (null model) is compared against a non-
parametric alternative (alternative model) at the level of latent variables. The
null extended model is obtained by incorporating the non Euclidean parameter
of the alternative model. This extension is de ned through a Bayesian Pseudo-
True Value, that makes the null model a reduction by suciency of the extended
model. The same observability process is introduced in both the null and the al-
ternative models; after integrating out the latent variables, a null and alternative
statistical models are accordingly obtained. The comparison is made between the
posterior measures of the non Euclidean parameter (of the alternative model) in
the extended and in the alternative statistical models. The general development
is illustrated with an example where only a linear combination of a latent vector
is observed; in the example, the partial observability is known up to the vector
de ning the observed linear combination. Some identi ability issues are treated
and the example shows the operationality and some pitfalls of the proposed test,
through a numerical experiment
关键词：Bayesian encompassing; Bayesian speci cation test; Dirichlet prior;
Partial observability.
