Internal regret in on-line portfolio selection

DOI10.1007/s10994-005-0465-4zbMath1111.91020OpenAlexW1963858753MaRDI QIDQ5916205

Gábor Lugosi, Gilles Stoltz

Publication date: 17 June 2005

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s10994-005-0465-4

zbMATH Keywords

universal portfoliointernal regretEG strategyindividual sequenceson-line investment

Mathematics Subject Classification ID

Applications of statistics to actuarial sciences and financial mathematics (62P05) Portfolio theory (91G10) Online algorithms; streaming algorithms (68W27)

Related Items

Replicator dynamics: old and new ⋮ A general internal regret-free strategy ⋮ Constrained no-regret learning ⋮ PAMR: passive aggressive mean reversion strategy for portfolio selection ⋮ Forecasting electricity consumption by aggregating specialized experts ⋮ Approachability, regret and calibration: implications and equivalences ⋮ Supermartingales in Prediction with Expert Advice ⋮ Unnamed Item ⋮ Sampled fictitious play is Hannan consistent ⋮ Note on universal conditional consistency ⋮ Online portfolio selection ⋮ How long to equilibrium? The communication complexity of uncoupled equilibrium procedures ⋮ A tree-weighting approach to sequential decision problems with multiplicative loss ⋮ Supermartingales in prediction with expert advice ⋮ Competitive Portfolio Selection Using Stochastic Predictions ⋮ Exponential weight approachability, applications to calibration and regret minimization

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• An analog of the minimax theorem for vector payoffs
• Potential-based algorithms in on-line prediction and game theory
• On prediction of individual sequences
• Adaptive and self-confident on-line learning algorithms
• Regret in the on-line decision problem
• Conditional universal consistency.
• Universal portfolios with and without transaction costs.
• Learning correlated equilibria in games with compact sets of strategies
• The Cost of Achieving the Best Portfolio in Hindsight
• An algorithm for maximizing expected log investment return
• Asymptotic calibration
• Universal Portfolios
• A Simple Adaptive Procedure Leading to Correlated Equilibrium
• On‐Line Portfolio Selection Using Multiplicative Updates
• Universal portfolios with side information
• Learning Theory and Kernel Machines
• A general class of adaptive strategies