[R] Testing a timeseries for heteroscedastic (lmtest)

[Achim Zeileis]

On Wed, 14 Jan 2004, Simon Ohrem wrote:

> Hello,
> 
> my intention is to test a given timeseries (vector) for heteroscedastic.
> Now I found the gqtest/bptest/hmctest in the lmtest package but a
> "formula" as argument is needed.
> What is meant by formula and how du I interpret the timeseries as
> formula.

lmtest is designed for testing linear regression models which are
typically described by a formula. The simplest conceivable model is
  x ~ 1
i.e. x (which can be a time series) is described as a mean plus noise. And
the above mentioned tests test whether the noise is homoskedastic or not.

The package lmtest also has a vignette telling you a bit more about how
these tests can be applied. Also note that in certain cases the tests in
the package strucchange and some diagnostic tests from tseries can be used
for testing for heteroskedasticity.

hth,
Z
 
> Thanks in advance
> 
> Simon
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>