# The local stability of convexity, affinity and of the Jensen equation

- 43 Downloads
- 8 Citations

## Summary.

Let \( C_D, A_D, J_D \) denote the smallest constants involved in the stability of convexity, affinity and of the Jensen equation of functions defined on a convex subset *D* of \( {\Bbb R}^n \). By a theorem of J. W. Green, \( C_D \le c\cdot \log (n+1) \) for every convex \( D\subset {\Bbb R}^n \), where *c* is an absolute constant. We prove that the lower estimate \( C_D \ge c\cdot \log (n+1) \) is also true, supposing that int \( D \neq {\not 0} \).¶We show that \( A_D \le 2 C_D \) and \( A_D \le J_D \le 2A_D \) for every convex \( D\subset {\Bbb R}^n \). The constant \( J_D \) is not always of the same order of magnitude as \( C_D \); for example \( J_D = 1 \) if \( D ={\Bbb R}^n \). We prove that there are convex sets (e.g. the *n*-dimensional simplex) with \( J_D \ge c\cdot \log n \).

## Keywords

Convex Subset Lower Estimate Local Stability Absolute Constant Small Constant## Preview

Unable to display preview. Download preview PDF.