Sample Question #85 (econometrics – time series)
What is cointegration? How do you test for cointegration? What’s an example of a cointegration application?
Cointegration refers to the econometric relationship between two or more time series that, while nonstationary with a unit root themselves, have a linear combination of the two that is stationary. More formally, let y be a vector time series (i.e., each component of y is a time series). While each y(i) is nonstationary with unit root, there exists some nonzero scalar vector a so that a’y is stationary.
Testing for cointegration is complicated. First, you should recall that the null hypothesis is there is no cointegration, i.e., all linear combinations of the components of y have unit root. Second, you should know there are two cases to consider, one is if the vector a has a known candidate value (e.g., as suggested by economic theory), and the other is if the vector a is totally unknown. Third, you need to know one testing procedure, the Engle-Granger two-step test.
In finance, there are many applications of cointegration. Many statistical arbitrage models, for instance, rely on the cointegration between two stocks’ prices. Another example is the S&P 500 index and its futures price.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
Notify me of new posts via email.