Moment inequalities for functions of independent random variables (Q1775439)

A number of rather general inequalities for functions of independent random variables is proved. In fact the authors suggest a method for obtaining moment inequalities. Their main and typical two inequalities are quoted: Let \(X_1,\dots, X_n\) be independent random variables with values in some measurable set \({\mathcal X},X_1',\dots, X_n'\) be independent copies of \(X_1,\dots,X_n\), \(F:{\mathcal X}^n\to\mathbb{R}\) be a measurable function, \(Z= F(X_1,\dots, X_n)\), \(Z_i'= F(X_1,\dots, X_{i-1}, X_i', X_{i+1},\dots, x_n),\) \[ V^+= E\Biggl[\sum^n_{i=1} (Z- Z_i')^2_+\,\mid\,X_1,\dots, X_n \Biggr],\quad V^-= E\Biggl[\sum^n_{i=1} (Z- Z_i')^2_-\,\mid\,X_1, \dots, X_n\Biggr], \] where \(V^\pm\) are conditional expectations with respect to the sigma-algebra generated by the random variables \(X_1,\dots, X_n\), \(x_+= \max\{x,0\}\), \(x_-= \max\{-x, 0\}\). Then \[ (E|(Z- EZ)_\pm|_q)^{2/q}\leq {(q-1)\sqrt{e}\over \sqrt{e-1}}(E(V^\pm)^{q/2})^{2/q},\quad q\geq 2. \]