RT Journal Article
T1 Dynamic generation of investment recommendations using grammatical evolution
A1 Martín Fernández, Carlos
A1 Quintana, David
A1 Isasi, Pedro
AB The attainment of trading rules using Grammatical Evolution traditionally follows a static approach. A singlerule is obtained and then used to generate investment recommendations over time. The main disadvantage ofthis approach is that it does not consider the need to adapt to the structural changes that are often associatedwith financial time series. We improve the canonical approach introducing an alternative that involves adynamic selection mechanism that switches between an active rule and a candidate one optimized for the mostrecent market data available. The proposed solution seeks the flexibility required by structural changes whilelimiting the transaction costs commonly associated with constant model updates. The performance of thealgorithm is compared with four alternatives: the standard static approach; a sliding window-based generationof trading rules that are used for a single time period, and two ensemble-based strategies. The experimentalresults, based on market data, show that the suggested approach beats the rest.
PB Universidad Internacional de la Rioja (UNIR)
SN 1989-1660
YR 2021
FD 2021-04-22
LK https://hdl.handle.net/10016/34521
UL https://hdl.handle.net/10016/34521
LA eng
NO The authors would like to acknowledge the financial support of the Spanish Ministry of Science, Innovation and Universities under grant PGC2018-096849-B-I00 (MCFin). This work has been supported by the Madrid Government (Comunidad de Madrid-Spain) under the Multiannual Agreement with UC3M in the line of Excellence of University Professors (EPUC3MXX), and in the context of the V PRICIT (Regional Programme of Research and Technological Innovation).
DS e-Archivo
RD 17 jul. 2024