[R] FW: lasso regression

[David Winsemius]

On Jul 12, 2011, at 10:12 AM, Heiman, Thomas J. wrote:

>
> Hi,
>
> Hopefully I got the formatting down.. I am trying to do a lasso  
> regression using the lars package with the following data (the data  
> files is in .csv format):
>
> 	V1		V2		V3			V4			V5		V6		V7		V8		V9
> 1	FastestTime	WinPercentage	PlacePercentage	ShowPercentage	 
> BreakAverage	FinishAverage	Time7Average	Time3Average	Finish
> 2	116.9		0.14285715	0.14285715		0.2857143		4.428571	3.2857144	 
> 117.557144	117.76667	5.0
> 3	116.22857	0.2857143	0.42857143		0.14285715		6.142857	2.142857	 
> 116.84286	116.8		2.0
> 4	116.41428	0.0		0.14285715		0.2857143		5.714286	3.7142856	117.24286	 
> 117.166664	4.0
> 5	115.8		0.5714286	0.0			0.2857143		2.142857	2.5714285	116.21429	 
> 116.53333	6.0
>

It is now clear that you failed to get your data in properly. Since  
stringsAsFactors is set to TRUE by default for all of the read.*  
functions, all of your columns are now factors. Perhaps you had a  
blank line at the beginning of your data? The default for read.csv  
(which is just a wrapper with different parameters for read.table) is  
to set header =TRUE. You should learn to use str() on your data  
immediately after data entry steps.

-- 
David.

> #load Data
> crs<- read.csv("file:///C:/temp/Horse//horseracing.csv<file:///C:\temp\Horse\horseracing.csv 
> >", na.strings=c(",", "NA", "", "?"), encoding="UTF-8")
>
> #    # define x and y
> x= x<-crs[,9]    #predictor variables
> y= y<-crs[1:8,]  #response variable
>
>
> library(lars)
> cv.lars(x, y, K=10, trace=TRUE, plot.it = TRUE,se = TRUE,  
> type="lasso")
>
> and I get:
>
> LASSO sequence
> Error in one %*% x : requires numeric/complex matrix/vector arguments
>
> Any idea on what I am doing wrong?  Thank you!!
>
> Sincerely,
>
> tom
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
West Hartford, CT