No label defined (Q4669409)

A strictly positive function \(f\) defined on a subinterval \(I\) of \((0,\infty)\) is said to be strictly multiplicatively convex if \(f(x^{1/2}y^{1/2} )<f(x)^{1/2}f(y)^{1/2}\) for all \(x,y\in I\) with \(x\neq y.\) It is called strictly multiplicatively concave if the reverse inequality holds true. The main result of the paper under review is Theorem 1, which asserts that the gamma function is strictly multiplicatively concave on \((0,x_{0}]\) and strictly multiplicatively convex on \([x_{0},\infty),\) where \(x_{0}=0.21609.\) is the unique solution of the equation \(\psi(x)+x\psi^{\prime}(x)=0.\) Here \(\psi(x)=\frac{d}{dx}\log\Gamma(x)\) denotes the digamma function. This answers a question raised by \textit{C. P. Niculescu} in his paper ``Convexity according to the geometric mean'' [Math. Inequal. Appl. 3, No. 2, 155--167 (2000; Zbl 0952.26006)]. A generalization concerning the convexity of the gamma function relative to Hölder's means is also presented. For the entire collection see [Zbl 1048.26003].