[R] Durbin-Watson test in packages "car" and "lmtest"

[ripley@stats.ox.ac.uk]

1) Achim pointed out that dwtest was one-sided and durbin.watson is
two-sided (from memory), so these are both indicators of close to
expectation.

2) Even if 0.44 and 0.92 were comparable, would it make the slightest
difference to your conclusions?  You have no evidence of serial
correlation either way.

3) Do worry about the big things first.  Your model (with the missing
coefficients supplied on the RHS, I presume) is an autoregression, and so
neither lm nor the Durbin-Watson test is entirely appropriate.  You should
be using functions such as gls (nlme) or arima0 (ts) to fit this model.
(It may be that least-squares will do a reasonable job, but it's always
suboptimal and the easiest way to find out by how much is to do the job
properly.  Things have changed since that 1951 paper.)

On Fri, 19 Apr 2002, Ahmad Abu Hammour wrote:

> Here is an example in which P-values have a significant difference.
  The data is a sub sample of a larger one.
> Basically the model I am using is as follows:
> y_{t+2}= y_{t+1} + y_t + I(x_{t+1}-z_{t+1}-w_{t+1}) +e_{t+2}

[...]

-- 
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 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._