Performance functions and reinforcement learning for trading systems and portfolios

Article Abstract:

A technique for training trading systems and portfolios that uses reinforcement learning algorithms to optimize objective functions that directly measure trading and investment performance is proposed. Performance functions such as profit and wealth, the Sharpe ratio and a proposed differential Sharpe ratio were optimized to test the methodology. Results affirm the efficacy of some of the proposed methods for optimizing trading systems and portfolios.

Methods, Management, Stocks, Portfolio management, Investments

---

The evolution of conventions

Article Abstract:

An analysis of the development of conventions and the role of expectations with more than one equilibrium condition is presented. The analysis develops a repeated n-game by various agents who draw individual strategies based on information about what other agents have done before. It is shown that a stochastic process which converges to a pure strategy Nash equilibrium can be developed. This process generates stochastically stable equilibria.

Analysis, Consumer behavior, Behavioral assessment, Markov processes

---

Bayesian learning behaviour and the stability of equilibrium forecasts

Article Abstract:

A learning model incorporating Bayesian characteristics and set in a stochastic environment is discussed. This model incorporates rational expectations about the state which involve forecasts and beliefs. These are updated continually and form part of a dynamic learning interaction. It is shown that full learning does not always happen. Nonetheless, partial learning has its results which come in the form of better informed predictions.

Models, Learning, Rational expectations (Economics), Teaching models

Similar abstracts:

• Abstracts: Performance of GARCH models in forecasting stock market volatility. Forecasting interest rate swap spreads using domestic and international risk factors: evidence from linear and non-linear models
• Abstracts: Local convergence of recursive learning to steady states and cycles in stochastic nonlinear models. The sequential equilibrium theory of reputation building: a further test
• Abstracts: Inequality in male and female earnings: the role of hours and wages. Capital gains taxes and realizations: evidence from interstate comparisons
• Abstracts: The Nash solutino and the utility of bargaining: a corrigendum. Randomization as an instrumental variable
• Abstracts: Financing constraints and inventory investment: a comparative study with high-frequency panel data. Capacity utilization under regulatory constraints

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.