Hi,
My experiment consist of factorial structure with 25 genotypes and 3
salinity levels ( 25 cross 3) each with 3 replications. 14 responses were
taken from the experiment and I applied MANOVA but I am applying
discriminant analysis to know which response variable has significant
impact on the separation of groups. I am doing the following:
MANOVA<-manova(cbind(RL,SL,RFW,SFW,RDW,SDW,R.S..DW.,LA,NL,Na,K,Ca)~geneotype*S.Level,data=data)
For discriminant analysis, I made a group of 1 genotype with one level of
Salinity level hence a total of 75 groups.
group<-paste("g", gl(75,3,length=225))
LDA<-lda(group~RL+SL+RFW+SFW+RDW+SDW+R.S..DW.+LA+NL+Na+K+Ca+K.Na+Ca.Na,data=Sdata)
Is it appropriate way? Moreover when I am plotting discriminant scores
means with the circle of radius r=z/sqrt(replication) but the graph doesn't
make any sense as the circles are too small too see.
Thanks,
Adeela
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lda.pdf
Type: application/pdf
Size: 5835 bytes
Desc: not available
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20150626/15ce5441/attachment.pdf>
Offtopic. This list is about R programming. Post to a statistics list like stats.stackexchange.com instead. Better yet, find a local expert to help you. What you describe sounds confused and likely to produce nonsense to me. You may not even have the information needed to answer the question you asked. Cheers, Bert Bert Gunter "Data is not information. Information is not knowledge. And knowledge is certainly not wisdom." -- Clifford Stoll On Thu, Jun 25, 2015 at 11:09 PM, adeela uaf <adeela.uaf at gmail.com> wrote:> Hi, > My experiment consist of factorial structure with 25 genotypes and 3 > salinity levels ( 25 cross 3) each with 3 replications. 14 responses were > taken from the experiment and I applied MANOVA but I am applying > discriminant analysis to know which response variable has significant > impact on the separation of groups. I am doing the following: > MANOVA<-manova(cbind(RL,SL,RFW,SFW,RDW,SDW,R.S..DW.,LA,NL,Na,K,Ca)~geneotype*S.Level,data=data) > For discriminant analysis, I made a group of 1 genotype with one level of > Salinity level hence a total of 75 groups. > group<-paste("g", gl(75,3,length=225)) > LDA<-lda(group~RL+SL+RFW+SFW+RDW+SDW+R.S..DW.+LA+NL+Na+K+Ca+K.Na+Ca.Na,data=Sdata) > > Is it appropriate way? Moreover when I am plotting discriminant scores > means with the circle of radius r=z/sqrt(replication) but the graph doesn't > make any sense as the circles are too small too see. > Thanks, > Adeela > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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.
Hi adeela,
I can't quite get what you want to plot in your example, but using the
iris data set and the example in the "lda" help page, this may give
you some idea of what can be done. I'm using the "z" structure in
the
example.
matplot(z$scaling,xlim=c(0,5),ylim=c(-3,3),pch="",axes=FALSE,
main="LDA scaling and means for iris data",ylab="Scaling",
xlab="Measurement")
axis(1,at=1:4,labels=c("Sepal.L.","Sepal.W.","Petal.L.","Petal.W."))
axis(2)
box()
library(plotrix)
draw.circle(rep(1:4,2),z$scaling,radius=rep(colMeans(z$means)/10,2),
col=c("azure","lightblue","dodgerblue","turquoise",
"tomato","red","darkred","indianred"))
text(c(3.5,3.5),c(-0.2,-0.6),c("Blues - LD1","Reds - LD2"))
Jim
On Fri, Jun 26, 2015 at 11:57 PM, Bert Gunter <bgunter.4567 at gmail.com>
wrote:> Offtopic. This list is about R programming. Post to a statistics list
> like stats.stackexchange.com instead. Better yet, find a local expert
> to help you. What you describe sounds confused and likely to produce
> nonsense to me. You may not even have the information needed to answer
> the question you asked.
>
> Cheers,
> Bert
> Bert Gunter
>
> "Data is not information. Information is not knowledge. And knowledge
> is certainly not wisdom."
> -- Clifford Stoll
>
>
> On Thu, Jun 25, 2015 at 11:09 PM, adeela uaf <adeela.uaf at
gmail.com> wrote:
>> Hi,
>> My experiment consist of factorial structure with 25 genotypes and 3
>> salinity levels ( 25 cross 3) each with 3 replications. 14 responses
were
>> taken from the experiment and I applied MANOVA but I am applying
>> discriminant analysis to know which response variable has significant
>> impact on the separation of groups. I am doing the following:
>>
MANOVA<-manova(cbind(RL,SL,RFW,SFW,RDW,SDW,R.S..DW.,LA,NL,Na,K,Ca)~geneotype*S.Level,data=data)
>> For discriminant analysis, I made a group of 1 genotype with one level
of
>> Salinity level hence a total of 75 groups.
>> group<-paste("g", gl(75,3,length=225))
>>
LDA<-lda(group~RL+SL+RFW+SFW+RDW+SDW+R.S..DW.+LA+NL+Na+K+Ca+K.Na+Ca.Na,data=Sdata)
>>
>> Is it appropriate way? Moreover when I am plotting discriminant scores
>> means with the circle of radius r=z/sqrt(replication) but the graph
doesn't
>> make any sense as the circles are too small too see.
>> Thanks,
>> Adeela
>>
>> ______________________________________________
>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> 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.
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.