A very important problem in time series analysis is testing for randomness against serial dependence. Classical parametric methods, commonly based on the autocorrelation coefficient, can be misleading when the underlying distribution assumption are not fulfilled. In this paper the use of a nonparametric measure of serial dependence, based on Gini's cograduation index, is discussed. This method is compared, in a simple application to financial data, with other nonparametric tests for randomness and with the classical autocorrelation coefficient.
Borroni, C. (2003). A nonparametric measure of autocorrelation in time series: an application to financial data. In Statistics in Social and Economic Applications: Proceeding of the International Conference held in Nizhny Novgorod 14-15 february 2003 (translated from Russian; original title: ПРИКЛАДНАЯ СТАТИСТИКА В СОЦИАЛЬНО-ЭКОНОМИЧЕСКИХ ПРОБЛЕМАХ) (pp.19-23). Nizhny Novgorod : State University of Nizhny Novgorod "N.I. Lobachevsky".
A nonparametric measure of autocorrelation in time series: an application to financial data
BORRONI, CLAUDIO GIOVANNI
2003
Abstract
A very important problem in time series analysis is testing for randomness against serial dependence. Classical parametric methods, commonly based on the autocorrelation coefficient, can be misleading when the underlying distribution assumption are not fulfilled. In this paper the use of a nonparametric measure of serial dependence, based on Gini's cograduation index, is discussed. This method is compared, in a simple application to financial data, with other nonparametric tests for randomness and with the classical autocorrelation coefficient.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
