hello... I am trying to get this code to work, but as I get to the predict
command, it displays an error due to the length of the data sets from the
removal of the NA's. Here is the data, and the code that I am using so
far, if you run it, you'll see the error pop up.... please help me to get
around this problem. Thanks in advance.
Location Dist. Size
low1 .5 10.5
low2 .5 23
low3 .5 NA
low4 .5 NA
mid1 3 15
mid2 3 11.5
mid3 3 15
mid4 3 NA
high1 6 12.5
high2 6 20
high3 6 21
high4 6 22
wall1 10 13
wall2 10 12.5
wall3 10 13
wall4 10 12
d<-read.table("dilomaaethiops.txt",header=T)
d
str(d)
range(d$Dist.)
s<-d$Size
range(s)
s[s<.1]<-NA
range(s,na.rm=T)
summary(d)
length(s)
length(d$Dist.)
plot(s~d$Dist.,xlab="Distance from Shoreline (m)",ylab="D. aethiops Size
(mm)", pch=7)
a1<-glm(s~d$Dist.,family="quasipoisson",data=d)
scatter.smooth(d$Dist.,s,ylab="D. aethiops Length (mm)",xlab="Distance from
Shoreline (m)",col="red")
summary(a1)
a2<-aov(s~d$Dist.,data=d)
summary(a2)
shapiro.test(a2$residuals)
qqnorm(a2$residuals)
qqline(a2$residuals)
fligner.test(s~d$Dist.,data=d)
kruskal.test(s~d$Dist.,data=d)
df<-df.residual(a1)
t<-qt(.975,df)
k<-data.frame(s=seq(10.5,max(s,na.rm=T),1))
p<-predict(a1,k,se=T)
est<-exp(p$fit)
low<-exp(p$fit-t*p$se.fit)
hight<-exp(p$fit-t*p$se.fit)
plot(d$Dist.,s,ylim=c(min(s),max(s)),xlab="Distance from Shorline (m)",
ylab="Length of D. aethiops (mm)", pch=7)
lines(k$s,est,lty=1,col="red")
lines(k$s,low,lty=2,col="blue")
lines(k$s,high,lty=2,col="blue")
[[alternative HTML version deleted]]
______________________________________________
R-help@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.
