The thesis dissertation is devoted to investigation of the performance of Relative Entropy Policy Search (REPS) reinforcement learning method in application to foreign exchange market trading. In order to accomplish this a software module was built in MATLAB environment, that is capable of backtesting of trading strategies as well as live trading with the same strategies with minimum adjustments. The parameter search using the REPS algorithm was performed for several benchmark trading strategies.
Application of reinforcement learning methods to foreign exchange market trading
ROMANOV, VLADIMIR
2014/2015
Abstract
The thesis dissertation is devoted to investigation of the performance of Relative Entropy Policy Search (REPS) reinforcement learning method in application to foreign exchange market trading. In order to accomplish this a software module was built in MATLAB environment, that is capable of backtesting of trading strategies as well as live trading with the same strategies with minimum adjustments. The parameter search using the REPS algorithm was performed for several benchmark trading strategies.File allegati
| File | Dimensione | Formato | |
|---|---|---|---|
|
Vladimir Romanov - Application of reinforcement learning methods to foreign exchange market trading - 2015 - Politecnico di Milano.pdf
non accessibile
Descrizione: Thesis text
Dimensione
3.09 MB
Formato
Adobe PDF
|
3.09 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
Utilizza questo identificativo per citare o creare un link a questo documento:
https://hdl.handle.net/10589/112810