A Markov-Bernstein type inequality for algebraic polynomials in \(L_{p}\), \(0<p<1\) (Q1908218)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A Markov-Bernstein type inequality for algebraic polynomials in \(L_{p}\), \(0<p<1\)
scientific article

    Statements

    A Markov-Bernstein type inequality for algebraic polynomials in \(L_{p}\), \(0<p<1\) (English)
    0 references
    26 June 1996
    0 references
    This note is devoted into proving a weighted Bernstein-Markov type inequality in \(L_p\), \(0<p<1\). Let \(\omega: \mathbb{R}^+\to \mathbb{R}^+\) satisfy \(M^{-1} \omega (t_1)\leq \omega (t_2)\leq M\omega (t_1)\), \(\forall t_1\leq t_2\leq 2t_1\), for some \(M>0\). Then with \(\rho_n (x):= n^{-1} (1- x^2)^{1/2}+ n^{-2}\), we have \[ \Biggl|{{P^{(r)} \rho^r_n} \over {\omega(\rho_n)}} \Biggr|_{L_p [-1,1 ]}\leq C\Biggl|{P\over {\omega (\rho_n)}} \Biggr|_{ L_p[0,1 ]}, \] where \(C= c^r_0 r!\) and \(c_0\) depends only on \(M\) and \(p\). The above inequality is known for \(\omega (t)= t^\alpha\), but the exact representation of the constant \(C\), which is important for obtaining inverse approximation theorems when \(0<p <1\), was known only in some cases like \(\omega (t) \equiv 1\).
    0 references
    Bernstein-Markov inequalities
    0 references
    0 references

    Identifiers