[R] Iterative estimation of linear regression model

Amarjit Singh Sethi set_alt at yahoo.co.in
Wed Oct 15 11:37:26 CEST 2008


Dear all
I am intrested in making iterative estimation (thro' loop statements) of, say, linear regression model. For this purpose, I have written the following programme and that I have made use of a sample data (viz., exp.txt):
 
Programme:
 
# Linear regression modelling with sample data (try5.txt)
# Repeated estimation through loop statement
x=read.table("try5.txt",header=T,sep="\t")
nm=names(x)
out="slrout.txt"
sink(out)
nvr=10;nv1=4;nv2=nvr-nv1
for(i in 1:nv1){
dep=nm[i+1]       
# The 1st dependent variable is C1 (at column no. 2)
for(j in 1:nv2){
ind=nm[j+nv1+1]   
# The first independent variable is C5 (at column no. 6)
# Estimation of simple linear regression equation 
slr=lm(dep ~ ind,data=x)
smr=summary(slr)
print(slr)
tsp=ts.plot(dep)
tsp
}
}
 
Data file (try.txt):
 
Year C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
Y1 3.5 13.8 9.5 6.8 0.4 24.0 3 7.1 24.6 0.03
Y2 3.8 13.9 9.9 7.6 0.7 12.8 5 9.0 32.5 0.01
Y3 4.5 14.5 14.2 9.2 0.6 14.5 1 7.5 22.9 0.05
Y4 5.9 16.2 24.6 12.7 0.2 24.3 3 8.0 26.8 0.04
Y5 7.2 20.4 40.6 18.2 0.8 28.2 5 9.2 44.3 0.02
Y6 5.9 18.6 37.4 14.5 0.3 36.9 8 9.5 32.9 0.10
Y7 8.0 16.1 88.6 24.1 0.1 34.6 2 8.7 11.1 0.02
Y8 13.6 21.1 56.3 19.0 0.7 33.3 6 9..5 10.8 0.06 
Y9 11.2 20.4 40.7 12.9 1.1 40.3 3 12.2 6.5 0.04
Y10 7.6 18.3 27.5 8.1 2.3 41.9 2 5.9 2.9 1.00
Y11 8.8 22.2 33.3 8.8 0.6 44.4 4 6.6 55.5 0.09
Y12 9.4 16.5 35.6 16.2 0.7 50.2 5 8..8 31.4 0.07
 
(These files have been given as attachments as well)

The data file has 10 variables in all, divided into two groups:  G1 consisting of the first 4 variables and G2 of the remaining 6 variables. We wish to iteratively pick one variable from the group G1 and regress it upon (an iteratively picked) variable from G2 and save the summary results. We also wish to save the repeatedly generated diagrams in an output file. But the programme has failed to work. Kindly help; I'm a raw hand in R-language.
 
Regards
ajss
 
 


      Be the first one to try the new Messenger 9 Beta! Go to http://in.messenger.yahoo.com/win/
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Linear Regression Model (through loop	statement).txt
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20081015/34f6e441/attachment.txt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: try.txt
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20081015/34f6e441/attachment-0001.txt>


More information about the R-help mailing list