[R] plotting polynomial regression line

Amit amitkumartiwary at gmail.com
Mon Dec 21 10:19:44 CET 2009


Thanks very much it solved my problem.

cheers,
Amit

On Sun, Dec 20, 2009 at 7:56 PM, David Winsemius <dwinsemius at comcast.net> wrote:
>
> On Dec 20, 2009, at 1:35 PM, Amit wrote:
>
>> Dear All,
>> I am trying to plot polynomial regression line to a scatterplot. I did
>> following so far:
>>
>>> x=c(1:9335)
>>> y=read.table("gp.txt",header=T,sep="\t")
>>> length(y$PCC) # y$PCC has values between 1 to 0 in decreasing order
>>
>> [1] 9335
>>>
>>> plot(x,y$PCC,col="red") #scatterplot between x and y$PCC
>>> reg=lm(y$PCC~poly(x,6)) # calculating polynomial fit with degree 6
>>> abline(reg,col="blue")
>>
>> Warning message:
>> In abline(reg, col = "blue") :
>>  only using the first two of 7regression coefficients
>>
>> After the above warning a line is drawn in the graph parallel to the
>> y-axis. But I was expecting a curve line through the scatterplot.
>> Am I doing something wrong? Please help!
>
> abline is designed to draw lines of the form y=a+bx, ... hence the name.
> (And as documented on its help page for regression objects).
>
> An effective method would be with lines and predict:
>
>  plot(cars, main="Stopping Distance versus Speed")
>  lines(cars$speed, predict(lm(dist~poly(speed, 6) ,data=cars) ), col="red")
>
>>
>> Best
>> Amit
>>
>> _______________
>
> David Winsemius, MD
> Heritage Laboratories
> West Hartford, CT
>
>




More information about the R-help mailing list