CORN: Correlation-Driven Nonparametric Learning Approach for Portfolio Selection -- an Online Appendix

TL;DR

This appendix provides a proof of the universal consistency of the CORN algorithm, a nonparametric method for portfolio selection, building on prior theoretical work and expert suggestions.

Contribution

It offers a rigorous proof of CORN's universal consistency, confirming its theoretical robustness in portfolio selection.

Findings

CORN is proven to be universally consistent.

The proof builds on Gy"orfi et al. [2006] and prior theoretical frameworks.

Expert validation supports the theoretical results.

Abstract

This appendix proves CORN's universal consistency. One of Bin's PhD thesis examiner (Special thanks to Vladimir Vovk from Royal Holloway, University of London) suggested that CORN is universal and provided sketch proof of Lemma 1.6, which is the key of this proof. Based on the proof in Gy\"prfi et al. [2006], we thus prove CORN's universal consistency. Note that the notations in this appendix follows Gy\"orfi et al. [2006].