An estimate of the root mean square error incurred when approximating an \(f\in L^2(\mathbb R)\) by a partial sum of its Hermite series (Q1649150)

Summary: Let \(f\) be a band-limited function in \(L^2(\mathbb R)\). Fix \(T>0\), and suppose \(f'\) exists and is integrable on \([-T,T]\). This paper gives a concrete estimate of the error incurred when approximating \(f\) in the root mean square by a partial sum of its Hermite series. Specifically, we show, that for \(K=2n\), \(n \in Z_+\), \(\left[ \frac{1}{2T},\int_{- T}^T,[f(t) -(S_K f)(t)]^2,d,t\right]^{1 / 2} \leq \left(1,+,\frac{1}{K}\right) \left(\left[\frac{1}{2 T},\int_{| t| > T},f,(t)^2,d,t\right]^{1/2},+,\left[\frac{1}{2 T},\int_{| \omega| > N},| \hat{f}(\omega)|^2,d,\omega\right]^{1/2}\right) + \frac{1}{K} \left[\frac{1}{2 T},\int_{| t| \leq T},f_N,(t)^2,d,t\right]^{1/2}+\frac{1}{\pi} \left(1,+,\frac{1}{2 K}\right) S_a(K, T),\) in which \(S_K f\) is the \(K\)-th partial sum of the Hermite series of \(f\), \(\hat{f}\) is the Fourier transform of \(f\), \(N=\frac{\sqrt{2K+1}+\sqrt{2K+3}}{2}\) and \(f_N = (\hat{f} \chi_{(- N, N)})^\vee(t) = \frac{1}{\pi} \int_{- \infty}^\infty \frac{\sin(N(t - s))}{t - s} f(s) d s\). An explicit upper bound is obtained for \(S_a(K, T)\).