[R] Help! - boxcox transformations

Peter Dunn dunn at usq.edu.au
Fri Dec 14 03:02:59 CET 2007

> I am trying to detrend and transform variables to achieve
> normality and stationarity (for time series use, namely spectral
> analysis). I am using the boxcox transformations.
> As my dataset contains zeros, I found I need to add a constant to
> it in order to run "boxcox". 

If your data are continuous with exact zeros, and contain a lot of 
zeros, no transformation may help.  You simply map a stack of zeros 
to some other value.  Box--Cox transforms are not really designed 
for this.

If your data really is continuous with exact zeros, other options 
exist such as hurdle models, Tweedie glms (package  tweedie) and so 


Dr Peter Dunn  |  dunn <at> usq.edu.au
Faculty of Sciences, USQ; http://www.sci.usq.edu.au/staff/dunn
Aust. Centre for Sustainable Catchments: www.usq.edu.au/acsc

This email (including any attached files) is confidentia...{{dropped:15}}

More information about the R-help mailing list