[R] adding trend to an arima model

Prof Brian Ripley ripley at stats.ox.ac.uk
Thu Mar 4 15:45:23 CET 2004


What does "+" give you?  It works for me -- simulate the model, simulate 
the trend and add them.

x <- arima.sim(list(ar=0.3), 100) + 1:100

is an AR(1) with a linear trend.

> arima(x, c(1,0,0), xreg=1:100)

Call:
arima(x = x, order = c(1, 0, 0), xreg = 1:100)

Coefficients:
         ar1  intercept   1:100
      0.4510    -0.1704  1.0031
s.e.  0.0915     0.3528  0.0060

...

On Thu, 4 Mar 2004, Samuel Kemp wrote:

> Does anyone know a method for adding a linear/polynominal trend to a 
> simulated arima model using the arima.sim function?

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595




More information about the R-help mailing list