Random function prediction and Stein's identity
Let X=(X1,X2,...,Xn) be a size n sample of i.i.d. random variables, whose distribution belong to the one-parameter ([theta]) continuous exponential family. We examine prediction functions of the form [theta]mh(X),m[greater-or-equal, slanted]1, where h is a polynomial in X. A natural identity that first appeared in Stein (Stein, 1973) and has been widely exploited since, is discussed in relation to members of such a family. Mild regularity conditions are also introduced that imply the nonexistence of a uniformly minimum mean squared error predictor for these functions.
Year of publication: |
2002
|
---|---|
Authors: | Nicoleris, Theodoros ; Sagris, Anthony |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 59.2002, 3, p. 293-305
|
Publisher: |
Elsevier |
Keywords: | Prediction Random function of a parameter Exponential family |
Saved in:
Saved in favorites
Similar items by person
-
A Flaming-Viot Process and Bayesian non Parametric
Nicoleris, Theodoros, (2006)
-
Parameter estimation for random dynamical systems using slice sampling
Hatjispyros, S.J., (2007)
-
A Bayesian nonparametric study of a dynamic nonlinear model
Hatjispyros, Spyridon J., (2009)
- More ...