Let ( Ω , A , P ) {\displaystyle (\Omega ,{\mathcal {A}},P)} be a probability space and Θ {\displaystyle \Theta } be a p ∈ N ∗ {\displaystyle p\in \mathbb {N} ^{*}} -dimensional parameter space. A set F := { F θ : θ ∈ Θ } {\displaystyle {\mathcal {F}}:=\{F_{\theta }:\theta \in \Theta \}} of probability distributions on A {\displaystyle {\mathcal {A}}} is called parametric model if ∀ B ∈ Σ , θ ∈ Θ ↦ F θ ( B ) {\displaystyle \forall B\in \Sigma ,\theta \in \Theta \mapsto F_{\theta }(B)} is well-defined.
