Cointegration and representation of integrated autoregressive processes in function spaces

12/23/2017
by   Won-Ki Seo, et al.
0

We provide a suitable generalization of cointegration for time series taking values in a potentially infinite dimensional Banach space or Hilbert space. Our main result is a generalization of the Granger-Johansen representation theorem: we obtain necessary and sufficient conditions for the existence of I(1) solutions to a given autoregressive law of motion, and a characterization of such solutions. To achieve this goal, we note that such an autoregressive law of motion is characterized by a simple linear operator pencil, so study its spectral properties. Our representation theorem requires a minimal assumption: we do not even assume compactness of autoregressive operators. Based on the results of our I(1) representation theorem, we also provide a representation theorem for I(2) autoregressive processes.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/12/2018

A note on Herglotz's theorem for time series on function spaces

In this article, we prove Herglotz's theorem for Hilbert-valued time ser...
research
09/26/2018

When is there a Representer Theorem? Reflexive Banach spaces

We consider a general regularised interpolation problem for learning a p...
research
03/16/2016

Short-term time series prediction using Hilbert space embeddings of autoregressive processes

Linear autoregressive models serve as basic representations of discrete ...
research
10/09/2020

Hilbert valued fractionally integrated autoregressive moving average processes with long memory operators

Fractionally integrated autoregressive moving average processes have bee...
research
08/20/2021

A Theoretical Analysis of the Stationarity of an Unrestricted Autoregression Process

The higher dimensional autoregressive models would describe some of the ...
research
07/12/2023

Stationarity with Occasionally Binding Constraints

This paper studies a class of multivariate threshold autoregressive mode...
research
08/09/2022

Partial autocorrelation parameterisation of models with unit roots on the unit circle

We propose a parametrization of autoregressive unit roots ARMA models (A...

Please sign up or login with your details

Forgot password? Click here to reset