Título:	Symbolic Time Series and Causality Detection: an Uneasy Alliance
Autor(es):	Contiggiani, Federico Eduardo Delbianco, Fernando Fioriti, Andrés Tohmé, Fernando
Fecha de publicación:	26-oct-2021
Revista:	IV JISC - Jornadas Interdisciplinarias en Sistemas Complejos 2021
Resumen:	Symbolic Time Series Analysis (STSA) is a quantitative dominant mixed method applied in Economics and other Social Sciences as a way of reducing the impact of noise on data and to exhibit more clearly the evolution of time series. We show that such transformation from numerical to symbolic series may fail to preserve relevant properties of the original series. We focus, in particular, on the existence of causal relations among series. Well known methods of detection of causality, like Transfer Entropy or Granger’s Test can either yield non-existing causal relations or miss some actually existing ones, which is highly relevant for a sound application of a mixed method like STSA.
URI:	http://rid.unrn.edu.ar/handle/20.500.12049/9551
Otros enlaces:	https://www.youtube.com/watch?v=HbKt_B4kyAs
Aparece en las colecciones:	Objetos de conferencia

Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons