How do you check if two variables are cointegrated?
Two sets of variables are cointegrated if a linear combination of those variables has a lower order of integration. For example, cointegration exists if a set of I(1) variables can be modeled with linear combinations that are I(0).
Are equity returns stationary?
No. Stock return is not always stationary. The solution to the problem is to transform the time series data so that it becomes stationary. If the non-stationary process is a random walk with or without a drift, it is transformed to stationary process by differencing.
Is it possible to check cointegration of variables that are integrated?
Ofcourse, these tests (ADF, PP or KPSS) are for detecting unit roots exclusively, but you must check your data for unit roots before you investigate if they are cointegrated or not. What’s the point of checking the dataset for cointegration if they’re not non-stationary in the first place? Remember, the drunk lady and her dog!
What’s the difference between EG and Johansen cointegration?
The difference between the two however, is that Johansen tests a system and can report more than one cointegrating vectors while EG tests only a single equation. Also, the precondition in both cases is that the variables must be integrated of order 1.
Can you use VAR model with Johansen cointegration test?
You can use var model even for 2 variables. The Johansen cointegration test helps you find any integration relationship in the long-run between variables. If you find cointegration vector by using your test, you have to apply Vector error correction approach. If you do not find any cointegration vector, then you have permission to use VAR model.
Is it possible to check the rank of cointegration?
The maximum number of rows of the beta’ matrix is the rank of cointegration, r, and the number of trends (or individual unit roots) driving the system is k-r. You may test the rank of cointegration of a system by means of the trace test or the lambda-max test of Johansen (1988).