Theorem 2 — the deterministic revenue bounds the optimal expected revenue: J*(n,t) ≤ J^D(n,t)
ProvedGVRPricing.FixedPrice.optValue_le_detValuedynamic-pricingfluid-approximationp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1revenue-management
Let be a regular demand function. For every stock and every horizon ,
where is the optimal expected revenue of the stochastic pricing problem (7) and the optimal value of the deterministic problem (11) with stock .
Uncertainty in sales can only lower the expected revenue. The bound gives every heuristic a computable benchmark, and Theorem 3 uses it to measure the fixed-price heuristic.
Formalization Note is the supremum over all non-anticipating policies, in , and is embedded in . The boundary cases and , which the page includes, are included.
Preamble
import Mathlib import Definitions.Def_GVRPricing_FixedPrice_Model import Definitions.Def_GVRPricing_FixedPrice_SalesProcess import Definitions.Def_GVRPricing_FixedPrice_Deterministic open MeasureTheory ProbabilityTheory Set open scoped ENNReal
Formal statement
namespace GVRPricing.FixedPrice
/-- **Theorem 2** (p. 1007). If `λ(p)` is a regular demand function, then for every stock
`0 ≤ n < ∞` and horizon `0 ≤ t < ∞`, `J*(n, t) ≤ J^D(n, t)`: the optimal expected revenue of
the stochastic problem (7) is at most the optimal value of the deterministic problem (11) with
`x = n`. -/
theorem optValue_le_detValue (M : Model) (n : ℕ) (t : ℝ) (ht : 0 ≤ t) :
optValue M n t ≤ ENNReal.ofReal (detValue M n t) := by sorry
end GVRPricing.FixedPrice
Source
Gallego, van Ryzin, Optimal Dynamic Pricing of Inventories with Stochastic Demand over Finite Horizons, Management Science 40(8) (1994), p. 1007 (PDF 9), §3.2.1, Theorem 2
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.