[R] Automated stepwise multiple linear regression

David Winsemius dwinsemius at comcast.net
Thu Jul 7 19:56:42 CEST 2011


On Jul 7, 2011, at 11:28 AM, klimator wrote:

> Dear forum members,
>
> hope you can understand my unprofessionell English. I have never been
> working with RStudio before (even not with R), so I immediately need  
> some
> help, because I want to carry out an automated multiple stepwise  
> linear
> regression between temperatures and different surface parameters.
> Data can be found in several .xls- or .csv-files within one folder.  
> Files
> are named "TMP_20090819_MLR_sfcparA" (data for August 19th 2009),  
> daywise
> going on until for example "TMP_20091231_MLR_sfcparA" (data for  
> December
> 31th 2009).
> Every single file contains a column "logger", "temperature",
> "surface_parameter1", "surface_parameter2", "surfaceparameter3", and  
> so on
> (all in all max. 10 surface parameters). Furthermore every data file  
> allways
> contains 21 datasets (=rows)).
>
> What I want R to do is to automtically open several files one after  
> another,
> to carry out multiple stepwise linear regression on 95%-statistical
> significance niveau between "temperature" and the independent  
> variables of
> all "surface_parameter(x)", to generate a new column to be filled  
> with the
> residua between "temperature" and model-estimation and finally to  
> save the
> output of the regression analysis to a correspondent file name  
> ("#.txt")
> into the same folder.
>
> Is anybody able to help me with my problem?

I'm sure there are, but it seems as though you are asking for someone  
to do a free consulting project or write a tutorial just for you.  
Perhaps someone will volunteer, but if they do not why don't you do  
some work on your own (after reading and working on "The Introduction  
to R" and the "R import/Export Manual")? The when you get stuck, ...  
and after reading the Posting Guide for what level of detail is  
requested, ...  submit another question that shows your efforts and is  
more focussed.

-- 
David Winsemius, MD
West Hartford, CT



More information about the R-help mailing list