Publication: Syncronicity between macroeconomic time series: an exploratory analysis
Loading...
Identifiers
Publication date
2000-05
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
In this paper we analyse the performances of a model free cointegration testing device that we
construct from functions of order statistics. First, we propose new exploratory techniques that
consist in comparing the plots of the range and the jump sequences of the original series. These
plots suggest alternative cointegration testing schemes. Here we focused on one of them, which
involves two complementary test statistics. We report on some promising results obtained from
Monte Carlo experiments as well as on some empirical applications of the new method to pairs of
exchange-rates, and to gold and silver prices. Our study concludes that the proposed methodology is
potentially robust to monotonic nonlinearities and serial correlation structure in the cointegration
errors, and certain types of level shifts in the cointegrating relationship.
Description
Keywords
Cointegration, Nonlinearity, Ranks, Ranges, Jumps