Dunkl-gamma type operators including Appell polynomials (Q2008160)

In this paper, the authors study the approximation properties of a new sequence of linear operators, defined in terms of the gamma function, the Dunkl generalization of the exponential function, and the Dunkel-Appell polynomials defined, in turn, from the Appell polynomials. The series of Appell polynomials \(\{ p_{k} (x) \}_{k=0}^\infty\) is defined by the formal equality \[ g (t) e^{xt} = \sum_{k = 0}^\infty p_{k} (x) t^{k} , \] where \(g(t) = \sum_{r=0}^\infty a_{r} t^{r}\) is a formal power series with complex coefficients \(a_{r}\), \(r = 0, 1 \dots\) and \(a_{0} \neq 0\). The Dunkl generalization of the exponential function, \(e_{\mu}(x)\), \(\mu >-1/2\), is defined by \[ e_{\mu}(x)=\sum_{k=0}^{\infty}\frac{x^k}{\gamma_{\mu}(k)}, \] where the coefficients \(\gamma_{\mu}\) are given by \[ \gamma_{\mu}(2k)=\frac{2^{2k}k!}{\Gamma (k+\mu +1/2)},\qquad \gamma_{\mu}(2k+1)=\frac{2^{2k+1}k!\Gamma (k+\mu +3/2)}{\Gamma (\mu +1/2)}. \] The Dunkl-Appell polynomials, also denoted in the paper by \(p_k(x)\), are generated by \[ A (t) e_{\mu}^{xt} = \sum_{k = 0}^\infty \frac{p_{k} (x)}{\gamma_{\mu}(k)} t^{k} , \] where \(A(t) = \sum_{r=0}^\infty \frac{a_{r}}{\gamma_{\mu}(r)} t^{r}\) is a formal power series with complex coefficients \(a_{r}\), \(r = 0, 1 \dots\) and \(a_{0} \neq 0\). Here it is the definition of the new operators, named as the Dunkl analogue of the Appell-Szász-Durrmeyer operators, which, as pointed out by the authors, arise naturally after a not short list of papers dealing with similar extensions of previous works. For \(x\in [0,\infty)\), \(f\in C[0,\infty)\), \[ D^*_n(f;x)=\frac{1}{e_{\mu}(nx)A(1)}\sum_{k=0}^{\infty}\frac{p_k(nx)}{\gamma_{\mu}(k)}\frac{n^{k+2\mu \theta_k+\lambda +1}}{\Gamma (k+2\mu \theta_k+\lambda +1)}\int_0^{\infty}e^{-nt}t^{k+2\mu \theta_k+\lambda}f(t)dt, \] where \(\mu ,\lambda \geq 0\), \(A(1)\ne 0\), \(\frac{a_{k-n}}{A(1)}\geq 0\), (\(0\leq n\leq k\)), \(k=0,1,2,\ldots\), and \(\theta_k=0\) if \(k=2p\), and \(\theta_k=1\) if \(k=2p+1\). The authors prove both qualitative and quantitative results for the approximation process of \(D^*_n(f;x)\) towards \(f(x)\). They follow a classical pattern, based on the use of the Korovkin-type theory, moduli of smoothness and the Peetre's K-functional.