Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/49193
Title: Dancing with bulls and bears: Nearest-neighbour forecasts for the Nikkei index
Authors: Fernández-Rodríguez, Fernando 
Sosvilla-Rivero, Simón
Dolores García-Artiles, María
Keywords: Efficient Capital-Markets
Nonlinear Dynamics
Stock Returns
Time-Series
Regression
Heteroskedasticity
Issue Date: 1999
Publisher: 0922-1425
Journal: Japan and the World Economy 
Abstract: In this paper we apply nearest-neighbour local predictors, inspired by the literature on forecasting in nonlinear systems, to the Nikkei 225 Index of the Tokyo Stock Market for the period 1 January 1986-5 June 1997. When forecasting performance is measured by Theil's U statistic, our nearest-neighbour predictors perform worse than a random walk, outperforming the random walk directional forecast. When formally testing for forecast accuracy, the results suggest that predictions from a random walk were statistically significantly better than the nearest-neighbour predictors for the entire forecasting period, as well as for one of the subperiods (a 'bull' market episode). Finally, when assessing the economic value of the nearest-neighbour predictors in absence of trading costs, the results of using them as a filter technique are superior to a buy-and-hold strategy for both the entire forecasting period acid for 'bear' market subperiods, where tests of 'forecast conditional efficiency' (or 'forecast encompassing') detected that the nearest-neighbour predictors contain useful information for forecasting the Nikkei Index that is not contained in the random walk. (C) 1999 Elsevier Science B.V. All rights reserved. JEL classification: C53; G15.
URI: http://hdl.handle.net/10553/49193
ISSN: 0922-1425
Source: Japan and the World Economy[ISSN 0922-1425],v. 11, p. 395-413
Appears in Collections:Artículos
Show full item record

SCOPUSTM   
Citations

27
checked on Apr 11, 2020

WEB OF SCIENCETM
Citations

17
checked on May 23, 2020

Page view(s)

4
checked on May 23, 2020

Google ScholarTM

Check


Share



Export metadata



Items in accedaCRIS are protected by copyright, with all rights reserved, unless otherwise indicated.