For any random variable \(X\) with finite expectation \(EX\) and for any convex function \(f\), the well-known Jensen's inequality \(f(EX) \leq Ef(X)\) holds, playing a significant role in probability and statistics theory. The aim of this note is to present an analogue of Jensen's inequality, where expectation is replaced by median. The author proves that Jensen's inequality for medians is satisfied by a class of functions that encloses the convex functions as a proper subset (the \(C\)-function class is very close to lower semi-continuous functions). The main tool for deriving the new type of inequality is a novel characterization of a median used as a new definition for the median, which entails also the natural extension of the Jensen's inequality to the higher-dimensional form (for multivariate distributions and multivariate \(C\)-functions).
- A refined Jensen's inequality in Hilbert spaces and empirical approximations
- Quantile Jensen's inequalities
- Generalized expectation with general kernels on g-semirings and its applications
- scientific article; zbMATH DE number 4171835 (Why is no real title available?)
- scientific article; zbMATH DE number 3953573 (Why is no real title available?)
- THE MEAN-MEDIAN-MODE INEQUALITY: COUNTEREXAMPLES
- Median aggregation of distribution functions
- Approximate calculation of Tukey's depth and median with high-dimensional data
- Construction-free median quasi-Monte Carlo rules for function spaces with unspecified smoothness and general weights
- Conditional quantiles: an operator-theoretical approach
- A Universal Median Quasi-Monte Carlo Integration
- On properties of fractional posterior in generalized reduced-rank regression
- Concentration properties of fractional posterior in 1-bit matrix completion
- Jensen's inequality for multivariate medians
This page was built for publication: Jensen's inequality for medians
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2483843)
