Re: [R] How to draw the decision boundaries for LDA and Rpart object
On Wed, 9 Aug 2006, Am Stat wrote: Dr. Ripley, R-help is not the address of `Dr. Ripley', nor even that of the person who wrote to you. Thanks very much for your help. I have used your partition tree and it works well. I am not familiar with the 'tree' package but I found that the threshold to make a cut returned by tree and rpart is almost the same value. Does that mean the decision boundaries for Rpart and Tree are the same for a same data when using the default value of parameters, no matter what the structure of data is? No, they can differ. There are comparisons in MASS (the book). Thanks very much! Leon - Original Message - From: Prof Brian Ripley [EMAIL PROTECTED] To: Am Stat [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Sent: Wednesday, August 09, 2006 3:04 AM Subject: Re: [R] How to draw the decision boundaries for LDA and Rpart object On Wed, 9 Aug 2006, Am Stat wrote: Hello useR, Could you please tell me how to draw the decision boundaries in a scatterplot of the original data for a LDA or Rpart object. There are examples in MASS (the book). For example: library(rpart) fit.rpart - rpart(as.factor(group.id)~., data=data.frame(Data) ) How can I draw the cutting lines on the orignial Data? Or is there any built in functions that can read the rpart object 'fit.rpart' to do that? See partition.tree in package tree. PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch 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. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch 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.
Re: [R] How to draw the decision boundaries for LDA and Rpart object
On Wed, 9 Aug 2006, Am Stat wrote: Hello useR, Could you please tell me how to draw the decision boundaries in a scatterplot of the original data for a LDA or Rpart object. There are examples in MASS (the book). For example: library(rpart) fit.rpart - rpart(as.factor(group.id)~., data=data.frame(Data) ) How can I draw the cutting lines on the orignial Data? Or is there any built in functions that can read the rpart object 'fit.rpart' to do that? See partition.tree in package tree. PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch 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.
Re: [R] How to draw the decision boundaries for LDA and Rpart object
Dr. Ripley, Thanks very much for your help. I have used your partition tree and it works well. I am not familiar with the 'tree' package but I found that the threshold to make a cut returned by tree and rpart is almost the same value. Does that mean the decision boundaries for Rpart and Tree are the same for a same data when using the default value of parameters, no matter what the structure of data is? Thanks very much! Leon - Original Message - From: Prof Brian Ripley [EMAIL PROTECTED] To: Am Stat [EMAIL PROTECTED] Cc: r-help@stat.math.ethz.ch Sent: Wednesday, August 09, 2006 3:04 AM Subject: Re: [R] How to draw the decision boundaries for LDA and Rpart object On Wed, 9 Aug 2006, Am Stat wrote: Hello useR, Could you please tell me how to draw the decision boundaries in a scatterplot of the original data for a LDA or Rpart object. There are examples in MASS (the book). For example: library(rpart) fit.rpart - rpart(as.factor(group.id)~., data=data.frame(Data) ) How can I draw the cutting lines on the orignial Data? Or is there any built in functions that can read the rpart object 'fit.rpart' to do that? See partition.tree in package tree. PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch 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.
Re: [R] How to draw the decision boundaries for LDA and Rpart object
Could you please tell me how to draw the decision boundaries in a scatterplot of the original data for a LDA or Rpart object. For example: library(rpart) fit.rpart - rpart(as.factor(group.id)~., data=data.frame(Data) ) You might want to have a look a classifly (http://had.co.nz/classifly/) which will do this in high dimensions. (You will need to install ggobi for it to work, however) Hadley __ R-help@stat.math.ethz.ch 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.
