NONPARAMETRIC KERNEL‐BASED SEQUENTIAL INVESTMENT STRATEGIES

169 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Cryptocurrency portfolio management with deep reinforcement learning (2016, background): ““Pattern-Matching” algorithms select part of history which is similar to current situation, and optimize the portfolio based on the selected history under some assumptions on the behavior of the market [6].”
Prediction, learning, and games (2006, background): “The assumption of independence was substantially relaxed by Algoet and Cover [6], who considered stationary and ergodic markets; see also Algoet [4,5], Walk and Yakowitz [304], Györfi and Schäfer [136], and Györfi, Lugosi, and Udina [135] for various results under such general assumptions.”
Portfolio Selection : A Survey (2014, unknown): “With different parameters (w, , or ρ), one can get a family of portfolios, which are then aggregated into a final portfolio using exponential weighting [Györfi et al. 2006].”
A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem (2017, background): “…Universal Portfolios (UP) (Cover, 1991), Exponential Gradient (EG) (Helmbold et al., 1998), Nonparametric Kernel Based Log Optimal Strategy (B K ) (Gy¨orﬁ et al., 2006), Correlation-driven Nonparametric Learning Strategy (CORN) (Li et al., 2011), and M0 (Borodin et al., 2000), except Weighted…”