This work concerns the optimization of a Trading Systems (TS) based on a small set of Technical Analysis (TA) indicators. Usually, in TA the values of the parameters (window lengths and thresholds) of these indicators are fixed by professional experience. Here, we propose to design the parametric configuration according to historical data, optimizing some performance measures subjected to proper constraints using a Particle Swarm Optimization-based metaheuristic. In particular, such an optimization procedure is applied to obtain both the optimal parameter values and the optimal weighting of the trading signals from the considered TA indicators, in order to provide an optimal trading decision. The use of a metaheuristic is necessary since the involved optimization problem is strongly nonlinear, nondifferentiable and mixed-integer. The proposed TS is optimized using the daily adjusted closing returns of seven Italian stocks coming from different industries and of two stock market indices.
Trading system mixed-integer optimization by PSO
Marco Corazza
;Francesca Parpinel
;Claudio Pizzi
2021-01-01
Abstract
This work concerns the optimization of a Trading Systems (TS) based on a small set of Technical Analysis (TA) indicators. Usually, in TA the values of the parameters (window lengths and thresholds) of these indicators are fixed by professional experience. Here, we propose to design the parametric configuration according to historical data, optimizing some performance measures subjected to proper constraints using a Particle Swarm Optimization-based metaheuristic. In particular, such an optimization procedure is applied to obtain both the optimal parameter values and the optimal weighting of the trading signals from the considered TA indicators, in order to provide an optimal trading decision. The use of a metaheuristic is necessary since the involved optimization problem is strongly nonlinear, nondifferentiable and mixed-integer. The proposed TS is optimized using the daily adjusted closing returns of seven Italian stocks coming from different industries and of two stock market indices.File | Dimensione | Formato | |
---|---|---|---|
2021-Corazza_Parpinel_Pizzi-Trading_system_mixed-integer_optimization_by_PSO-eMAF2020.pdf
non disponibili
Descrizione: Articolo nella versione dell'editore.
Tipologia:
Versione dell'editore
Licenza:
Accesso chiuso-personale
Dimensione
5.76 MB
Formato
Adobe PDF
|
5.76 MB | Adobe PDF | Visualizza/Apri |
I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.