TheoremBase

Below-Frontier Evaluations, Tilted Poisson Moment Functions, and Multi-Base Window Bounds for Jointly Driven Solutions of the Controlled N-Agent Dynamics

Statement

Adopt the setting, notation, and hypotheses of Level-Revealed Conditioning for Jointly Driven Solutions of the Controlled N-Agent Dynamics --- in particular the σ\sigma-algebra S0\mathcal{S}_0 and the independence of the finite family consisting of S0\mathcal{S}_0 together with the σ\sigma-algebra generated by the variables of each single clock --- together with the derived objects of Frontier-Window and Crossing-Compensation Identities for Jointly Driven Solutions of the Controlled N-Agent Dynamics (the fixed level rr, the caps cac_a, and the σ\sigma-algebra and events of the third paragraph of Level-Revealed Conditioning for Jointly Driven Solutions of the Controlled N-Agent Dynamics are not used in the present lemma): the probability space (Ω,F,P)(\Omega,\mathcal{F},P) with transition and observation clocks, the clock labels aa with clocks YaY^{a} and rate bounds BaB_a, the JJ given solutions of the controlled NN-agent dynamics on [0,T][0,T] with consumed clock times AjaA_j^{a}, for t∈[0,T]t\in[0,T] the σ\sigma-algebra Ft\mathbb{F}_t generated by the union of the JJ system σ\sigma-algebras at tt, and for each clock label aa the frontier consumed time At∨,a=max⁡(A1a(t),…,AJa(t))A^{\vee,a}_t=\max(A_1^{a}(t),\dots,A_J^{a}(t)). Let μp\mu_p and exp⁡\exp be as in Predictable-Window Moment Identities for the Homogeneous Poisson Process, and write E\mathbb{E} for the expectation, expectations of [0,∞][0,\infty]-valued random variables being their integrals with respect to PP. For a natural number p≥0p\ge0 and a real κ≥0\kappa\ge0 define μp,κ:[0,∞)→[0,∞]\mu_{p,\kappa}:[0,\infty)\to[0,\infty] by μp,κ(x) := exp⁡(−x) ∑k≥0kp exp⁡(κk) xkk!,\mu_{p,\kappa}(x)\ :=\ \exp(-x)\,\sum_{k\ge0}\frac{k^{p}\,\exp(\kappa k)\,x^{k}}{k!}, the sum over the natural numbers k≥0k\ge0, with the factorial k!k!, well defined by nonnegativity as the least upper bound of its finite partial sums, with the conventions 00=10^{0}=1 and k0=1k^{0}=1; all products of [0,∞][0,\infty]-valued quantities below are formed with the convention that zero times infinity is zero.

1. (Below-frontier evaluations) Let t∈[0,T]t\in[0,T], let aa be a clock label, and let WW be an Ft\mathbb{F}_t-measurable random variable with 0≤W(ω)≤At∨,a(ω)0\le W(\omega)\le A^{\vee,a}_t(\omega) for every ω\omega. Then the evaluation YWaY^{a}_{W} of the clock path at the level WW is an Ft\mathbb{F}_t-measurable random variable.

2. (Tilted product windows) Let t∈[0,T]t\in[0,T], let a1,…,aka_1,\dots,a_k be distinct clock labels with k≥1k\ge1 a natural number, and let Vˉ\bar{V} and Λˉ\bar{\Lambda} be nonnegative reals. For each i∈{1,…,k}i\in\{1,\dots,k\} let viv_i and λi\lambda_i be Ft\mathbb{F}_t-measurable random variables with At∨,ai(ω)≤vi(ω)≤VˉA^{\vee,a_i}_t(\omega)\le v_i(\omega)\le\bar{V} and 0≤λi(ω)≤Λˉ0\le\lambda_i(\omega)\le\bar{\Lambda} for every ω\omega, let pi≥0p_i\ge0 be a natural number and κi≥0\kappa_i\ge0 a real, and set Vi:=Yvi+λiai−YviaiV_i:=Y^{a_i}_{v_i+\lambda_i}-Y^{a_i}_{v_i}, a random variable with values in the natural numbers by claim 1 of Predictable-Window Moment Identities for the Homogeneous Poisson Process. Then for every Ft\mathbb{F}_t-measurable Z:Ω→[0,∞]Z:\Omega\to[0,\infty], in [0,∞][0,\infty] and with the finite product notation: E[Z ∏i=1k(Vipi exp⁡(κiVi))] = E[Z ∏i=1kμpi,κi(λi)].\mathbb{E}\Bigl[Z\,\prod_{i=1}^{k}\bigl(V_i^{p_i}\,\exp(\kappa_i V_i)\bigr)\Bigr]\ =\ \mathbb{E}\Bigl[Z\,\prod_{i=1}^{k}\mu_{p_i,\kappa_i}(\lambda_i)\Bigr].

3. (Moment functions) For every real κ≥0\kappa\ge0 and every x≥0x\ge0: μ0,κ(x)=exp⁡(x (exp⁡(κ)−1))\mu_{0,\kappa}(x)=\exp\bigl(x\,(\exp(\kappa)-1)\bigr), and μp,κ(x)\mu_{p,\kappa}(x) is finite for every natural number p≥0p\ge0. For every natural number p≥1p\ge1 and every x≥0x\ge0, μp,0(x)=μp(x)\mu_{p,0}(x)=\mu_p(x); and for every natural number p≥1p\ge1, every real κ≥0\kappa\ge0, every real Λˉ≥0\bar{\Lambda}\ge0, and every x∈[0,Λˉ]x\in[0,\bar{\Lambda}], μp,κ(x) ≤ cp,κ(Λˉ) x,cp,κ(Λˉ) := ∑k≥1kp exp⁡(κk) Λˉk−1k! < ∞.\mu_{p,\kappa}(x)\ \le\ c_{p,\kappa}(\bar{\Lambda})\,x,\qquad c_{p,\kappa}(\bar{\Lambda})\ :=\ \sum_{k\ge1}\frac{k^{p}\,\exp(\kappa k)\,\bar{\Lambda}^{k-1}}{k!}\ <\ \infty.

4. (Multi-base window bound) Let n≥1n\ge1 be a natural number, let 0≤t1≤⋯≤tn≤T0\le t_1\le\dots\le t_n\le T, let Z:Ω→[0,∞]Z:\Omega\to[0,\infty] be Ft1\mathbb{F}_{t_1}-measurable, and let Vˉ,Λˉ≥0\bar{V},\bar{\Lambda}\ge0 be reals. Let L\mathfrak{L} be a finite, possibly empty, set of distinct clock labels --- the fraktur L\mathfrak{L} being distinct from the control set A\mathcal{A} of the adopted setting --- and assign to each a∈La\in\mathfrak{L}: a base index b(a)∈{1,…,n}b(a)\in\{1,\dots,n\}; an Ftb(a)\mathbb{F}_{t_{b(a)}}-measurable anchor vav_a with Atb(a)∨,a(ω)≤va(ω)≤VˉA^{\vee,a}_{t_{b(a)}}(\omega)\le v_a(\omega)\le\bar{V} for every ω\omega; an Ftb(a)\mathbb{F}_{t_{b(a)}}-measurable width λa\lambda_a and a real λˉa∈[0,Λˉ]\bar{\lambda}_a\in[0,\bar{\Lambda}] with 0≤λa(ω)≤λˉa0\le\lambda_a(\omega)\le\bar{\lambda}_a for every ω\omega; a natural number pa≥0p_a\ge0; and a real κa≥0\kappa_a\ge0. Set Va:=Yva+λaa−YvaaV_a:=Y^{a}_{v_a+\lambda_a}-Y^{a}_{v_a}, a random variable with values in the natural numbers by claim 1 of Predictable-Window Moment Identities for the Homogeneous Poisson Process. Then, in [0,∞][0,\infty]: E[Z ∏a∈L(Vapa exp⁡(κaVa))] ≤ (∏a∈Lca) E[Z],\mathbb{E}\Bigl[Z\,\prod_{a\in\mathfrak{L}}\bigl(V_a^{p_a}\,\exp(\kappa_a V_a)\bigr)\Bigr]\ \le\ \Bigl(\prod_{a\in\mathfrak{L}}\mathfrak{c}_a\Bigr)\,\mathbb{E}[Z], where ca:=2(pa+1)nexp⁡(n λˉa(exp⁡(κa)−1))\mathfrak{c}_a:=2^{(p_a+1)n}\exp\bigl(n\,\bar{\lambda}_a(\exp(\kappa_a)-1)\bigr) if pa=0p_a=0, and ca:=2(pa+1)nexp⁡(n λˉa(exp⁡(κa)−1)) cpa,κa(Λˉ) λˉa\mathfrak{c}_a:=2^{(p_a+1)n}\exp\bigl(n\,\bar{\lambda}_a(\exp(\kappa_a)-1)\bigr)\,c_{p_a,\kappa_a}(\bar{\Lambda})\,\bar{\lambda}_a if pa≥1p_a\ge1.

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…