Discussion Paper
No. 2016-33 | July 05, 2016
Ignacio Arbués, Ramiro Ledo and Mariano Matilla-García
Automatic Identification of General Vector Error Correction Models
(Published in Recent Developments in Applied Economics)

Abstract

There are a number of econometrics tools to deal with the different type of situations in which cointegration can appear: I(1), I(2), seasonal, polynomial, etc. There are also different kinds of Vector Error Correction models related to these situations. We propose a unified theoretical and practical framework to deal with many of these situations. To this aim: (i) a general class of models is introduced in this paper and (ii) an automatic method to identify models, based on estimating the Smith form of an autoregressive model, is provided. Our simulations suggest the power of the new proposed methodology. An empirical example illustrates the methodology.

JEL Classification:

C01, C22, C32, C51, C52

Links

Cite As

[Please cite the corresponding journal article] Ignacio Arbués, Ramiro Ledo, and Mariano Matilla-García (2016). Automatic Identification of General Vector Error Correction Models. Economics Discussion Papers, No 2016-33, Kiel Institute for the World Economy. http://www.economics-ejournal.org/economics/discussionpapers/2016-33


Comments and Questions



Anonymous - Congratulations and codes demand
July 06, 2016 - 09:03
Congratulations for this good paper. Are there gauss codes for this study? And can I take these codes if it is possible? Thank you.

Mariano Matilla-Garcia - Code
September 05, 2016 - 12:38
Thank you for your comment on the paper. We have prepared all codes in Matlab. They will be freely available eventually. Currently we don´t have the code en Gauss

Anonymous - Codes
September 06, 2016 - 21:48
Hi Dear Mariano, Thank you for your reply. Where I can reach this matlab codes? Please send me its link. I will cite to your study. Sincerely...

Anonymous - Referee Report
September 13, 2016 - 08:33
The paper aims at setting up a unified theoretical framework that nests many multivariate time series structures used in econometric analysis. In this sense, it is a contribution in line with the works by Franchi (2006, 2007, cited in the paper). Although the paper brings interesting insights to the topic, it is unnecessarily long and dispersed, at times confusing, and would benefit from some streamlining. In particular: •One can move directly from the Wold representation in page 4 to proposition 2 in page 5, where the Smith form is presented. There is no need to spell out the Granger representation theorem beyond referring to it briefly, this result should be well known to the reader of this paper. •Although the structure of the first part of the paper, based on assumptions, propositions, remarks and examples, is probably aimed at making the paper easier to approach, it actually makes the paper more difficult to read and feels like an unnecessary slicing. I would recommend to get rid of the “Remark” and “Example” headers altogether and integrate these parts in the text in a narrative manner. •Either the proofs of propositions should be either integrated in the main text or the text should be more explicit in terms of referring to the appendix to find the proof of the respective propositions. •I do not know whether the review of the theory about Smith forms starting in page 10 is really useful. Referring to the necessary results in Hungerford (1980) when necessary would be probably more efficient. •The paper is in need of a concluding section where the analysis is summarized in a concise way.