Abstract
We study a functional equation whose unknown maps a Euclidean space into the space of probability distributions on [0,1]. We prove existence and uniqueness of its solution under suitable regularity and boundary conditions, we show that it depends continuously on the boundary datum, and we characterize solutions that are diffuse on [0,1]. A canonical solution is obtained by means of a Randomly Reinforced Urn with different reinforcement distributions having equal means. The general solution to the functional equation defines a new parametric collection of distributions on [0,1] generalizing the Beta family.
Similar content being viewed by others
References
Aletti, G., May, C., Secchi, P.: On the distribution of the limit proportion for a two-color, randomly reinforced urn with equal reinforcement distributions. Adv. Appl. Probab. 39(3), 690–707 (2007)
Aletti, G., May, C., Secchi, P.: A central limit theorem, and related results, for a two-color randomly reinforced urn. Adv. Appl. Probab. 41(3), 829–844 (2009)
Beggs, A.: On the convergence of reinforcement learning. J. Econ. Theory 122(1), 1–36 (2005)
Durham, S.D., Flournoy, N., Li, W.: A sequential design for maximizing the probability of a favourable response. Can. J. Stat. 26(3), 479–495 (1998)
Durham, S.D., Yu, K.F.: Randomized play-the-leader rules for sequential sampling from two populations. Probab. Eng. Inf. Sci. 4(03), 355–367 (1990)
Gibbs, A.L., Su, F.E.: On choosing and bounding probability metrics. Int. Stat. Rev. 70(3), 419–435 (2002)
Kumaraswamy, P.: A generalized probability density function for double-bounded random processes. J. Hydrol. 46(1–2), 79–88 (1980)
Li, W., Duhram, S., Flournoy, N.: Randomized Pólya urn designs. In: Proceedings of the Biometric Section of the American Statistical Association, pp. 166–170 (1996)
May, C., Flournoy, N.: Asymptotics in response-adaptive designs generated by a two-color, randomly reinforced urn. Ann. Stat. 37(2), 1058–1078 (2009)
May, C., Paganoni, A.M., Secchi, P.: On a two-color generalized Pólya urn. Metron 63(1), 115–134 (2005)
Muliere, P., Paganoni, A.M., Secchi, P.: A randomly reinforced urn. J. Stat. Plan. Inference 136(6), 1853–1874 (2006)
Paganoni, A.M., Secchi, P.: A numerical study for comparing two response-adaptive designs for continuous treatment effects. Stat. Methods Appl. 16(3), 321–346 (2007)
Wooff, D.A.: Bounds on reciprocal moments with applications and developments in Stein estimation and post-stratification. J. R. Stat. Soc. B 47(2), 362–371 (1985)
Acknowledgements
The authors thank two anonymous referees whose useful comments helped to clarify some key points of the paper.
Author information
Authors and Affiliations
Corresponding author
Appendix: Doob’s Decomposition of the RRU Process
Appendix: Doob’s Decomposition of the RRU Process
This appendix provides a series of auxiliary results necessary to prove Propositions 3.1 and 3.2. We will refer to the notations introduced in Sect. 3. For n=1,2,…, let \(\mathcal{A}_{n}=\sigma(\delta _{1},R_{X}(1),R_{Y}(1),\ldots,\delta_{n},R_{X}(n),R_{Y}(n) )\) and consider the filtration \(\{\mathcal{A}_{n}\}\); then, given the initial urn composition \((x,y) \in{\mathbb{S}}\), the Doob’s semi-martingale decomposition of Z n (x,y) is
where {M n } is a zero mean martingale and the previsible process {A n } is eventually increasing (decreasing), again by [11, Theorem 2]. We also denote by {〈M〉 n } the bracket process associated with {M n }, i.e. the previsible process obtained by the Doob’s decomposition of \(M^{2}_{n}\).
We first provide some auxiliary inequalities. As a consequence of [2, Lemma 4.1], we can bound the increments ΔA n of the Z n -compensator process and the increments Δ〈M〉 n of the bracket process associated with {M n }. In fact, an easy computation gives
and
where
and
Now, [2, Lemma 4.2] with \(m=\int_{0}^{\beta}k\mu (dk)=\int_{0}^{\beta}k\nu(dk)\) gives
By applying [2, Lemma 4.1] with \(h(x,t) = (\frac{x}{x+t})^{2}\), B D =[2β,∞), D=D n , R=R X (n+1) or R=R Y (n+1) and \({\mathcal{A}}={\mathcal{A}}_{n}\), one obtains:
on the set {D n ≥2β}. Since
if D 0≥2β, and thus β+D n ≥3β, (25) together with (26) yields
Lemma A.1
For all k=1,2,…,
If, in addition, D 0≥2β then, for all k,n=1,2,… ,
when d≥c≥0 and b k =c+D n −β+mk.
Proof
Let η ∗ be a random variable independent of \({\mathcal{A}}_{\infty}\) and let η 1 be a random variable independent of \(\sigma({\mathcal{A}}_{\infty},\eta^{*})\) and such that η 1/β has distribution Binomial(1,m/β). Define \({\mathcal{A}}_{k+n^{-}}^{*}=\sigma(\eta^{*},{\mathcal{A}}_{k+n-1},\mathbb{I}(k+n))\); by [2, Lemma 4.1], if D>0 is \({\mathcal{A}}_{k+n^{-}}^{*}\)-measurable and 0≤R≤β with \({\mathbb{E}}(R)=m\) is independent of \({\mathcal{A}}_{k+n^{-}}^{*}\), one obtains
and thus
Therefore, for c≥0, by applying (30) k times, we get
where η k is independent of \(\sigma({\mathcal{A}}_{\infty})\) and η k /β has distribution Binomial(k,m/β). Equation (28) is now a consequence of [13, (21)]: if \(\tilde{\eta_{k}}\sim\textrm{Binomial}(k,r)\) and l>0, then
Apply this to (31) with n=0, \(\tilde{\eta_{k}}=\eta_{k}/\beta\), l=D 0/β and r=m/β to obtain (28).
Equation (29) is a consequence of [13, (25)]: if \(\tilde{\eta_{k}}\sim\textrm{Binomial}(k,r)\) and l>1, then
Apply this to (31) with \(\tilde{\eta_{k}}=\eta_{k}/\beta\), l=D n +c/β (which is greater than 2) and r=m/β to obtain
Jensen’s inequality yields \({\mathbb{E}}((d+D_{n+k})^{-1} |{\mathcal{A}}_{n} ) \geq(d+D_{n}+mk)^{-1}\), and thus
Since
we get (29):
□
The following Lemmas A.2 and A.3 provide inequalities which control the previsible and the martingale part of the process Z n , respectively; they require that the initial composition of the urn is sufficiently large.
Lemma A.2
If D 0≥2β, then
Proof
Apply (29) with n=0, c=m, d=β. Equation (25) then reads
if b k =km+D 0−(β−m). Since A 0=0,
where the last inequality is true because β−m≤β≤D 0/2. □
Lemma A.3
Let D 0≥2β. For all n≥0,
Proof
Since Z n+k (1−Z n+k )≤1/4, by (27), one gets
Apply (29) with c=0 and d=β, obtaining
if b k =km+D n −β. Thus
since 2(D n −β)≥2(D 0−β)≥D 0. □
Rights and permissions
About this article
Cite this article
Aletti, G., May, C. & Secchi, P. A Functional Equation Whose Unknown is \(\mathcal{P}([0,1])\) Valued. J Theor Probab 25, 1207–1232 (2012). https://doi.org/10.1007/s10959-011-0399-7
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10959-011-0399-7