Characterization theorems for mean value insurance premium calculation principle (Q390617)

Let \(X\) be a random variable representing the size of insurance compensation related to a particular insurance contract, and let \(\pi(X)\) be the premium to be paid for the risk \(X\). There are many premium calculation prin\-cip\-les. One of them is the mean value premium principle, which is based on some auxiliary function \(v\in C_2(\mathbb{R})\) such that \(v'(x)>0\) and \(v''(x)\geq 0\) for all \(x\in \mathbb{R}\). Mean value premium \(\pi(X)\) for the risk \(X\) is the solution of the equation \(v(\pi(X))=\mathbb{E}(v(X))\) with the function \(v\) from above. The authors consider additivity, consistency, iterativity and scale invariance of the mean value premium principle. The results are formulated in a form of necessary and sufficient conditions for the auxiliary function \(v\). The following assertion is typical for the paper. The mean value premium calculation principle possesses the additivity property, i.e. \(\pi(X_1+X_2)=\pi(X_1)+\pi(X_2)\) for any pair of independent risks \(X_1\) and \(X_2\), if and only if \(v(x)=ax+b\) with \( a>0\) or \(v(x)=\alpha e^{\beta x}+\gamma\) with \( \min\{\alpha,\beta\}>0\).