Dear all
I need some help once again.s-plus "finmetrics" contains the standard
Dickey-Fuller unit root test.I tried to run the test for a matrix of
observations - a matrix which contains several time series in its columns.
The result is a critical probability value that
1) is different from the result obtained from the application of the test
on every single vector of the matrix
2) although, according to the test, every single vector is not
covariance-stationary, the result for the whole matrix points to
stationarity.
The question is: am I allowed to use this test for multivariate time series
- a matrix of observations, or the test is only for univariate time series,
namely vectors?
Do I have to read the result of the test as the variables forming the
matrix are cointegrated?
Thank you all
Nicola Beghetto
Product and Business Development
Banca IMI - Milan
Nicola Beghetto
-
This message is confidential and solely for the intended addressee(s). If
you are not the intended recipient of this message, please notify the
sender immediately and delete it from your system. Unauthorised
reproduction, disclosure, modification and or distribution of this e-mail
is strictly prohibited. The contents of this e-mail do not constitute a
commitment by Banca di Intermediazione Mobiliare IMI S.p.A. (Banca IMI),
except where expressly provided for in a written agreement between you and
Banca IMI.
Banca di Intermediazione Mobiliare IMI S.p.A. is an authorised Bank in
Italy.
Banca di Intermediazione Mobiliare IMI S.p.A., London Branch, a member of
the London Stock Exchange, is regulated by the Financial Services Authority
for the conduct of designated investment business in the UK
|