Asymptotic analysis of non-periodical cointegration with high seasonals

Asymptotic analysis of non-periodical cointegration with high seasonals

Cointegration tests are studied for seasonal time series with large periods. The study is performed asymptotically as the period tends to infinity, which is shown to allow a simple limit distribution to be established. These results offer a simple diagnostic tool for cointegration when the seasonal...

Full description

Saved in:
Bibliographic Details
Published in:Boletín de la Sociedad Matemática Mexicana Vol. 25; no. 2; pp. 443 - 459
Main Authors: Dickey, David A., González-Farías, Graciela, Muriel, Nelson
Format: Journal Article
Language:English
Published: Cham Springer International Publishing 01-07-2019
Springer Nature B.V
Subjects:
Asymptotic properties
Cointegration analysis
Diagnostic software
Diagnostic systems
Mathematics
Mathematics and Statistics
Original Article
Outbreaks
Time series
Nonstationarity
62M10
37M10
Cointegration
Periodicity
Asymptotic approximation
91B84
62F03
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cointegration tests are studied for seasonal time series with large periods. The study is performed asymptotically as the period tends to infinity, which is shown to allow a simple limit distribution to be established. These results offer a simple diagnostic tool for cointegration when the seasonal period is expected to be large, and its critical values do not depend on the form of the deterministic components present in the model. An application to climatic factors and outbreaks of dengue in Mexico illustrates the methods.
ISSN:1405-213X
2296-4495
DOI:10.1007/s40590-018-0201-2