Throughout, Path=Path(E,T), and for p∈Path we call the intervals [0,t1), [tj,tj+1) (1≤j<k) and [tk,T] of the definition (the single interval [0,T] when k=0) the constancy intervals of p; they partition [0,T], and each of them, except possibly the last, has its right endpoint strictly larger than its left endpoint and excluded, while the last one is [tk,T] or [0,T]. We use that the family of subsets of a set whose preimage under a given map lies in a given σ-algebra is itself a σ-algebra (preimages commute with complements and countable unions), so that a map into a generated σ-algebra is measurable as soon as the preimages of the generators are measurable. Arithmetic, Absolute Values, and Pointwise Limits of Measurable Real-Valued Functions is used for constants, indicators, linear combinations, products and pointwise limits of measurable real-valued maps.
Claim 1. Every subset S of the finite set E is the finite union of the singletons {y}, y∈S, so the preimage of S under p↦p(u) is the finite union of the generators {p:p(u)=y}, y∈S, a member of CT; thus the evaluation map is measurable. Since ϕ(p(u))=∑y∈Eϕ(y)1{p(u)=y}, the map p↦ϕ(p(u)) is a linear combination of indicators of members of CT, hence measurable. Now let 0<T′≤T and p∈Path with times t1<⋯<tk; let k′ be the number of indices j with tj≤T′. The restriction p∣[0,T′] is constant on [0,t1) if k′≥1, on [tj,tj+1) for j<k′, and on [tk′,T′], the last interval being contained in [tk′,tk′+1) if k′<k and in [tk,T] if k′=k; if k′=0 it is constant on [0,T′], which is contained in [0,t1) (or in [0,T] when k=0). Hence p∣[0,T′]∈Path(E,T′) with the times t1,…,tk′. The preimage under the restriction map of a generator {p′:p′(u)=y} of CT′ (u∈[0,T′]) is the generator {p:p(u)=y} of CT, so the restriction map is measurable.
Claim 2. Let p∈Path with times t1<⋯<tk and let t∈(0,T]. If k=0 or t≤t1, then p is constant on [0,t), and δ=t together with the constant value works. Otherwise let j be the largest index with tj<t; then p is constant on [tj,tj+1)⊇[tj,t) if j<k (as t≤tj+1), and on [tk,T]⊇[tk,t) if j=k; so δ=t−tj and the constant value work. If y and y′ both have the property with δ and δ′, then y=p(s)=y′ for s=t−min(δ,δ′)∈[0,t); so p(t−) is unique. For u∈[0,T), the constancy interval containing u has right endpoint strictly larger than u (if it is the last one, its right endpoint is T>u and δ=T−u works; otherwise it is of the form [a,b) with u∈[a,b), so b>u and it contains [u,u+δ) for δ=b−u), which gives the right-continuity assertion. For measurability, fix t∈(0,T] and y∈E and let n0 be a natural number with 1/n0≤t. We claim
{p:p(t−)=y}=n≥n0⋃ m≥n⋂{p:p(t−1/m)=y}.
If p(t−)=y with the δ above, then for every m≥max(n0,1/δ) the point t−1/m lies in [t−δ,t), so p(t−1/m)=y, and p belongs to the right side. Conversely, if p(t−1/m)=y for all m≥n, then, with δ as in the existence part, p(t−1/m)=p(t−) for all m≥max(n,n0,1/δ), so p(t−)=y. The right side is a countable union of countable intersections of generators, hence a member of CT; as in claim 1 this gives the measurability of p↦p(t−).
Claim 3. Fix y∈E. For a natural number n and i∈{1,…,n} put In,i=((i−1)T/n,iT/n], a member of B[0,T], and define gn:[0,T]→[0,T] by gn(0)=0 and gn(u)=iT/n for u∈In,i; the sets {0} and In,1,…,In,n partition [0,T], and u≤gn(u)<u+T/n for every u. Then
1{p(gn(u))=y}=1{0}(u)1{p(0)=y}+i=1∑n1In,i(u)1{p(iT/n)=y}((u,p)∈[0,T]×Path),
where 1I denotes the indicator of a set I; each summand is the indicator of a measurable rectangle of B[0,T]⊗CT, so the left side is B[0,T]⊗CT-measurable as a function of (u,p). For every (u,p) the sequence 1{p(gn(u))=y} converges to 1{p(u)=y}: if u=T then gn(T)=T for every n; if u<T, claim 2 gives δ>0 with p=p(u) on [u,u+δ)∩[0,T], and for n>T/δ we have gn(u)∈[u,u+δ), so p(gn(u))=p(u). Being a pointwise limit of measurable maps, (u,p)↦1{p(u)=y} is B[0,T]⊗CT-measurable.
Now let ϕ:E→R and put M=maxy∈E∣ϕ(y)∣. The map (u,p)↦ϕ(p(u))=∑y∈Eϕ(y)1{p(u)=y} is B[0,T]⊗CT-measurable and bounded by M in absolute value. For fixed p its section u↦ϕ(p(u)) is B[0,T]-measurable by claim 3 of Sections of Product-Measurable Sets and Maps Are Measurable, and Insertion Maps into Products Are Measurable, and, being bounded, it is integrable over [0,T]: by claim 4 of Restricted Lebesgue Measure and Integral Toolkit on a Compact Interval (applied with the second function equal to the constant 1), a B[0,T]-measurable f with ∫[0,T]f2du<∞ has integrable absolute value, and here f2≤M2 gives ∫[0,T]f2du≤M2∫[0,T]1du=M2T by claim 1 of Linearity and Monotonicity of the Lebesgue Integral, the integral of the constant 1 being λ[0,T]([0,T])=T by Simple Function and Its Integral and claim 1 of Restricted Lebesgue Measure and Integral Toolkit on a Compact Interval; so f, and in particular u↦ϕ(p(u)), is integrable. By claim 2 of Linearity and Monotonicity of the Lebesgue Integral, ∣∫[0,T]ϕ(p(u))du∣≤∫[0,T]∣ϕ(p(u))∣du≤∫[0,T]Mdu=MT, and, the indicators u↦1{p(u)=y} being integrable by the same argument,
∫[0,T]ϕ(p(u))du=y∈E∑ϕ(y)∫[0,T]1{p(u)=y}du.
It remains to see that p↦∫[0,T]1{p(u)=y}du is CT-measurable for each y. For each n and each p, the map u↦1{p(gn(u))=y} is the nonnegative simple function 1{0}(u)1{p(0)=y}+∑i=1n1In,i(u)1{p(iT/n)=y} on [0,T], whose integral is ∑i=1n(T/n)1{p(iT/n)=y} by additivity (claim 1 of Linearity and Monotonicity of the Lebesgue Integral) and Simple Function and Its Integral, the integral of c1I being cλ[0,T](I), with λ[0,T]({0})=0 and λ[0,T](In,i)=T/n by claim 1 of Restricted Lebesgue Measure and Integral Toolkit on a Compact Interval and Existence of Lebesgue Measure on the Real Line; this is a CT-measurable function of p, being a linear combination of indicators of generators. For fixed p, the maps u↦1{p(gn(u))=y} are B[0,T]-measurable, converge pointwise on [0,T] to u↦1{p(u)=y} as shown above, and are dominated by the constant 1, which is integrable over [0,T] since λ[0,T]([0,T])=T<∞; by the dominated convergence theorem their integrals converge to ∫[0,T]1{p(u)=y}du. Hence p↦∫[0,T]1{p(u)=y}du is a pointwise limit of CT-measurable maps, so it is CT-measurable, and the display shows that p↦∫[0,T]ϕ(p(u))du is a linear combination of measurable maps. This completes the proof.