Let , where has the standard normal distribution and . Suppose is increasing, concave and twice continuously differentiable, with sufficient integrability to differentiate its Gaussian expectations twice and integrate by parts. A convenient sufficient bound is for fixed constants . For ,The middle equality uses Gaussian integration by parts. If , the mean derivative is strictly positive. At , the scale derivative is zero. Allowing negative leaves the distribution unchanged but reverses the sign of that derivative, so the monotonicity assertion is stated on nonnegative scale.
Furthermore,is negative semidefinite: its quadratic form on is . Thus is jointly concave in mean and scale. Concavity also follows directly by applying concavity of pointwise to affine combinations of , without assuming the differentiability formulas when their integrability fails.
Articles by others on the same topic
There are currently no matching articles.