[R-SIG-Finance] Seasonal Dummy Variables in VECM (tsDyn)

mtrokic mirza.trokic at gmail.com
Mon Sep 26 22:47:43 CEST 2011


I am currently working on the VECM chapters covering Lutkepohl's "New intro
to multiple time series analysis". I would like to use the tsDyn library to
replicate his examples. However, I for the life of me cannot understand how
to include seasonal dummy variables in tsDyn:VECM. 

The time series Lutkepohl uses is bivariate quarterly data from 1972 Q2 to
1998 Q4. The procedure I would like to use is: 


e6ts = read.table('e6.dat', header = TRUE) 
e6ts = ts(e6ts, start = c(1972,2), frequency=4) 

e6ts.vecm.ml = VECM(e6ts, lag=3, r=1, include='const', estim='ML', LRinclude
= ???) 

It is here where I get stuck. I want to include seasonal dummy variables of
the sort D = (1,s1,s2,s3)'. However, I have no clue how to set up this
matrix so that I can include it with LRinclude. In fact, whenever I try to
put any matrix for LRinclude, such as LRinclude = D, it tells me that it
must be NULL or a character vector. 

Can someone please be so kind and lead me though performing this example. I
really need to learn how to do this. 

Thanking you in advance. 

Mirza T.

View this message in context: http://r.789695.n4.nabble.com/Seasonal-Dummy-Variables-in-VECM-tsDyn-tp3845323p3845323.html
Sent from the Rmetrics mailing list archive at Nabble.com.

More information about the R-SIG-Finance mailing list