[R] Confusion about cointegration for AR(M) model

Zhou, Cindy Cindy.Zhou at citizensbank.com
Mon Mar 23 21:07:13 CET 2015


I have a question regarding the concept of cointegration. Does the concept of cointegration apply to any model? Or it only applies to OLS? For example, I fit an autoregressive error model , AR(M)

y_t=x_t*â+v_t

v_t=-ö_1*v_(t-1)-...-ö_m*v_(t-m)+å_t

If the ADF tests prove that both dependent variable y and independent variable x are stationary at I(1), and the residuals å_t are stationary at I(0), Can I conclude y and x are cointegrated? The residuals å_t are the prediction errors from the AR(M) model.

I checked the definition of cointegration, it looks like as long as the residuals from OLS are I(0) and both x and y are I(1), then x and y are cointegrated? Do I need to test stationary of the residuals from my AR(M) model in order to prove that x and y are cointegrated?

You help is appreciated.




Use of email is inherently insecure. Confidential information, including account information, and personally identifiable information, should not be transmitted via email, or email attachment. The information in this email may contain confidential and/or privileged information and is intended only for the use of the individual/entity named above. Any disclosure, copying, distribution or use of this information is strictly prohibited. If you have received this communication in error, please notify the sender immediately and destroy any record of this email.

Citizens Bank, N.A. is an affiliate of Citizens Financial Group, Inc.

	[[alternative HTML version deleted]]



More information about the R-help mailing list