Hi everyone!
I am trying to make two log-normal AR(0,1) model using R with a given
correlation between them, \rho, on the form:
X_t = \alpha X_{t-1} + a_t
Y_t = \beta Y_{t-1} + b_t
At the moment I have been making n values of correlated log-normal data,
called a_t and b_t, and generated a starting value X[1] and Y[1] using the
rnorm() function. The rest of the n-1 values are calculated in a for() loop.
The data do get a lognormal "look", but it is obviously not a
lognormal
distribution.
As I am a novice to time-series, my question is simply: Are there any way to
make correlated log-normal distributed AR(0,1) models, and are there any
package in R that will help me?
sincerely
Chris
--
View this message in context:
http://r.789695.n4.nabble.com/Lognormal-AR-0-1-model-tp4688176.html
Sent from the R help mailing list archive at Nabble.com.