We construct an error space which is smaller than but admits analogous existence theorems for optimal predictor schemes.

Results

Construction

Given we define to be the set of functions s.t. . It is easily seen is an error space.

Given , denote . We define to be the set of bounded functions s.t. for any , if then

We define .

Proposition 1

is an error space for any . is an error space.

Proposition 2

Consider a polynomial . There is a function s.t.

(i)

(ii) For any function we have


The proofs of Propositions 1 and 2 are in the Appendix. The following are proved using exactly like the analogous statements for and we omit the proofs.

Lemma

Consider a distributional estimation problem, , -predictor schemes. Suppose a polynomial and are s.t.

Then s.t.

Theorem 1

Consider a distributional estimation problem. Define by

Define by

Then, is a -optimal predictor scheme for .

Theorem 2

There is an oracle machine that accepts an oracle of signature and a polynomial where the allowed oracle calls are for and computes a function of signature s.t. for any , a distributional estimation problem and a corresponding -generator, is a -optimal predictor scheme for .

Appendix

Proof of Proposition 1

The only slightly non-obvious condition is (v). We have

Proof of Proposition 2

Given functions s.t. for , the proposition for implies the proposition for by setting

Therefore, it is enough to prove to proposition for functions of the form for .

Consider any . We have

Since takes values in

Similarly

The last two equations imply that

Raising to a power is equivalent to adding a constant to , therefore

Since we can choose satisfying condition (i) so that

It follows that

Personal Blog

0

New Comment