Asymptotic behavior of posterior distributions for random processes under incorrect models
In this paper, the asymptotic behavior of posterior distributions on parameters contained in random processes is examined when the specified model for the densities is not necessarily correct. Uniform convergence of likelihood functions in some way is shown to be a sufficient condition for the posterior distributions to be asymptotically confined to a set (Theorem 1). For ergodic stationary Markov processes uniform convergence of likelihood functions is established by the ergodic theorem for Banach-valued stationary processes (Proposition 1). A sufficient condition for the uniform convergence is also shown for general random processes (Proposition 2).
These results are used to analyze the asymptotic behavior of posterior distributions on parameters contained in linear systems under incorrect models (Example 1 and 2).
- R.H. Berk, Limiting behavior of posterior distributions when the model is incorrect, Ann. Math. Statist., 37 (1966), 51–58Google Scholar
- I.I. Gikhman and A.V. Skorphod, "Introduction to the Theory of Random Processes," W.B. Sounders Co., Philadelphia (1969)Google Scholar
- S. Kullback, "Information and Statistics," Wiley, New York (1959)Google Scholar
- W.J. Padgett and R.L. Taylar, "Laws of Large Numbers for Normed Linear Spaces and Certain Fréchet Spaces," Lecture Notes in Mathematics, Springer-Verlag, Berlin (1973)Google Scholar
- K.R. Parthasarathy, "Probability Measures on Metric Spaces," Academic Press, New York (1967)Google Scholar
- M. Aoki and P.C. Yue, On certain convergence questions in system identification, SIAM. J. Control, 8 (1970)Google Scholar
- R.W. Brockett, "Finite Dimensional Linear Systems," Wiley, New York (1970)Google Scholar
- K. Yoshida, "Functional Analysis," Academic Press, New York (1968)Google Scholar