Cost functions and model combination for VaR-based asset allocation using neural networks

IEEE Transactions on Neural Networks

Volumn 12, Issue 4, 2001, Pages 890-906

  Chapados, Nicolas ^a,b   Bengio, Yoshua ^b  

^a IEEE   (Canada)

^b UNIVERSITÉ DE MONTRÉAL   (Canada)

Author keywords

Asset allocation; Financial performance criterion; Model combination; Recurrent multilayer neural networks; Value at risk

Indexed keywords

ASSET ALLOCATION; COST FUNCTIONS; MODEL COMBINATION;

DECISION MAKING; FINANCE; PERFORMANCE; RESOURCE ALLOCATION;

RECURRENT NEURAL NETWORKS;

EID: 0035391774     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/72.935098     Document Type: Article

References (25)

1. Training a neural network with a financial criterion rather than a prediction criterion
2. Nonlinear trading models through sharpe ratio maximization
3. Optimization of trading systems and portfolios
4. Finding optimal portfolios with constraints on value at risk
5. Asset allocation in a value-at-risk framework
6. Neural networks for density estimation
7. Optimization criteria for learning algorithms in portfolio selection
8. The sharpe ratio
9. Learning internal representations by error propagation
10. Learning translation invariant in massively parallel networks
11. Backpropagation, weight-elimination and time series prediction
12. Tracking the best expert
13. Additive versus exponentiated gradient updates for linear prediction
14. A reality check for data snooping
15. Quantiles