AFTD Auto-Formalizing Theoretical Domains

bivariate_poisson_pmf_nonneg

theoremverified

For any natural numbers a and b, 0 ≤ bivariate_poisson_pmf a b.

Statement

theorem bivariate_poisson_pmf_nonneg (a b : ℕ) :
    0 ≤ bivariate_poisson_pmf a b
Source
step towards dual_melas_adjacent_cyclic_patterns_bivariate_poisson_limit
Verified
03 Nov 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 1
  • bivariate_poisson_pmfThe bivariate Poisson probability mass function bivariate_poisson_pmf a b is defined by Real.exp (-(3/2)) * ∑ j ∈ Finset.range (min a b +…

The proof, explained

The nonnegativity of the bivariate Poisson probability mass function follows directly from its definition. Unfolding the definition reveals a product of an exponential factor and a finite sum of quotients of powers of $1/2$ by factorials. Because every constituent building block—the exponential, the power, and the factorials—is strictly positive, the expression is nonnegative, which Lean's `positivity` tactic verifies automatically.

  1. It suffices to show that $\exp(-3/2) \sum_{j=0}^{\min(a, b)} \frac{(1/2)^{a+b-j}}{j! (a-j)! (b-j)!} \ge 0$.

    Unfolding the definition of `bivariate_poisson_pmf` replaces the PMF with its explicit formula: the product of the constant factor $\exp(-3/2)$ and a finite sum over $j \in \{0, \dots, \min(a, b)\}$ of terms $(1/2)^{a+b-j} / (j!(a-j)!(b-j)!)$.

    line 4

      unfold bivariate_poisson_pmf
  2. $0 \le \text{bivariate\_poisson\_pmf } a\ b$ automation

    The `positivity` tactic discharges the goal by inspecting the signs of all components: the exponential $\exp(-3/2)$ is strictly positive, $(1/2)^{a+b-j}$ is positive, the factorials of natural numbers are positive, quotients of positive numbers are positive, and any sum of nonnegative terms is nonnegative.

    line 5

      positivity

Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.

Lean source view module on GitHub

/-- The bivariate Poisson PMF is nonnegative everywhere. -/theorem bivariate_poisson_pmf_nonneg (a b : ℕ) :    0 ≤ bivariate_poisson_pmf a b := by  unfold bivariate_poisson_pmf  positivity
See it in the graphDiscuss this resultChallenge it