An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20111128/193ec18c/attachment.pl>
Comparing data
3 messages · Matevz Pavlic, Florent D.
Unlike a data.frame, a matrix can only hold one type of data. Since
you have a column of characters ("labSt") in your original data,
turning it into a matrix will give you a matrix of characters. You can
check it is the case by asking class(DF.m[1,1]).
So you'll have to remove this labSt column before you use as.matrix,
maybe use it for your rownames if you want to keep labSt around.
On Mon, Nov 28, 2011 at 11:49 AM, Matev? Pavli?
<matevz.pavlic at gi-zrmk.si> wrote:
Hi all,
i have a data set cintaining 14 columns ?and 11 rows. Rows represent single point and columns represent the parameter measured.
I wiuld like to compare the data to see which are more alike. I used the cluster analysis, but now i ma wondering if there are some other methods, since the cluster analysis did not give me the result i like.
I tried to use factanal() for factor analysis but i geta an error saying that it can only be used on numerical data (which is odd since all i have is numerical data).
For factanal i used this code :
DFm<-read.csv("kemijaM.csv", sep=";", dec=".") ? ? ? ? ? # input data
DF.m<-as.matrix(DFm) ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? # transform it to matrix
factanal(DF, factors=5) ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?# factor analysis
Here is what data looks like....
'data.frame': ?11 obs. of ?14 variables:
$ labSt ? ? ? ? ? ?: Factor w/ 11 levels "2011/9496","2011/9497",..: 1 2 3 4 5 6 7 8 9 10 ...
$ Mangan ? ? ? ? ? : num ?0.1313 0.6364 0.0889 0.0028 0 ...
$ Kalcij ? ? ? ? ? : num ?0.192 0.566 0.172 0 0.303 ...
$ Magnezij ? ? ? ? : num ?0.321 0.568 0.29 0.352 0 ...
$ Fluorid ? ? ? ? ?: num ?0.6739 0.2283 0.7826 0.0109 0.2609 ...
$ Hidrogenkarbonati: num ?0.362 1 0.31 0 0.16 ...
$ Kalij ? ? ? ? ? ?: num ?0.335 0.456 0.335 0.426 0 ...
$ Sulfat ? ? ? ? ? : num ?0.692 0 0.517 0.609 0.609 ...
$ Nitrati ? ? ? ? ?: num ?0.3934 0.0195 0.5008 0.3835 0.2601 ...
$ Silikati ? ? ? ? : num ?0.55 1 0.5214 0.0429 0.7143 ...
$ Fe ? ? ? ? ? ? ? : num ?0.786 0 0.357 0.214 0 ...
$ Natrij ? ? ? ? ? : num ?0.1321 0.093 0.093 0.0579 0 ...
$ m.Alkaliteta ? ? : num ?0.362 1 0.31 0 0.161 ...
$ Klorid ? ? ? ? ? : num ?0.0372 0.02471 0.03895 0.02721 0.00107 ...
Any ideas would be greatly appreciated. ? ? ? ?[[alternative HTML version deleted]]
______________________________________________ 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.
Forgot to suggest: if labSt is the 1st column in your file, you could just add "row.names = 1" to your read.csv call.
On Mon, Nov 28, 2011 at 8:06 PM, Florent D. <flodel at gmail.com> wrote:
Unlike a data.frame, a matrix can only hold one type of data. Since
you have a column of characters ("labSt") in your original data,
turning it into a matrix will give you a matrix of characters. You can
check it is the case by asking class(DF.m[1,1]).
So you'll have to remove this labSt column before you use as.matrix,
maybe use it for your rownames if you want to keep labSt around.
On Mon, Nov 28, 2011 at 11:49 AM, Matev? Pavli?
<matevz.pavlic at gi-zrmk.si> wrote:
Hi all,
i have a data set cintaining 14 columns ?and 11 rows. Rows represent single point and columns represent the parameter measured.
I wiuld like to compare the data to see which are more alike. I used the cluster analysis, but now i ma wondering if there are some other methods, since the cluster analysis did not give me the result i like.
I tried to use factanal() for factor analysis but i geta an error saying that it can only be used on numerical data (which is odd since all i have is numerical data).
For factanal i used this code :
DFm<-read.csv("kemijaM.csv", sep=";", dec=".") ? ? ? ? ? # input data
DF.m<-as.matrix(DFm) ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? # transform it to matrix
factanal(DF, factors=5) ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?# factor analysis
Here is what data looks like....
'data.frame': ?11 obs. of ?14 variables:
$ labSt ? ? ? ? ? ?: Factor w/ 11 levels "2011/9496","2011/9497",..: 1 2 3 4 5 6 7 8 9 10 ...
$ Mangan ? ? ? ? ? : num ?0.1313 0.6364 0.0889 0.0028 0 ...
$ Kalcij ? ? ? ? ? : num ?0.192 0.566 0.172 0 0.303 ...
$ Magnezij ? ? ? ? : num ?0.321 0.568 0.29 0.352 0 ...
$ Fluorid ? ? ? ? ?: num ?0.6739 0.2283 0.7826 0.0109 0.2609 ...
$ Hidrogenkarbonati: num ?0.362 1 0.31 0 0.16 ...
$ Kalij ? ? ? ? ? ?: num ?0.335 0.456 0.335 0.426 0 ...
$ Sulfat ? ? ? ? ? : num ?0.692 0 0.517 0.609 0.609 ...
$ Nitrati ? ? ? ? ?: num ?0.3934 0.0195 0.5008 0.3835 0.2601 ...
$ Silikati ? ? ? ? : num ?0.55 1 0.5214 0.0429 0.7143 ...
$ Fe ? ? ? ? ? ? ? : num ?0.786 0 0.357 0.214 0 ...
$ Natrij ? ? ? ? ? : num ?0.1321 0.093 0.093 0.0579 0 ...
$ m.Alkaliteta ? ? : num ?0.362 1 0.31 0 0.161 ...
$ Klorid ? ? ? ? ? : num ?0.0372 0.02471 0.03895 0.02721 0.00107 ...
Any ideas would be greatly appreciated. ? ? ? ?[[alternative HTML version deleted]]
______________________________________________ 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.