similar to: Mapping PCA loadings on to map

Hello, Following on from a paper by Ludwig, Horel and Whiteman in 2004, I have been using PCA to identify surface wind patterns. In the paper they approach the problem using the "real vector approach", that is, they create paired (u,v) vectors for the wind data, in effect artificially doubling the number of variables, then perform the pca, subsequently repairing the output and plotting

Hi, I am applying the following code to map pca loadings onto a 3d grid, my problem is this - the output only plots the spheres in the requested color (in this case "red") for the first argument. The sphere from the second argument appear as flat dark circles. Also the text3d() command only seems to work for a couple of the positions, with no text added in most cases. Could anyone offer

Hi, I'm having a mental block as to how I can automatically assign filenames to the output of the following code. I am wishing to create a separate .png file for every image created, each of them having a sequential filename ie "sourcefile_index.png" so that I can create a movie from them. Please could someone tell me where I am going wrong? the following code works fine and

Please could someone point me in the right direction as I appear to be having a total mental block with fairly basic PCA problem! I have a large dataframe where rows represent independent observations and columns are variables. I am wanting to perform PCA sequentially on blocks of nrows at a time and produce a graphical output of the loadings for the first 2 EOFs for each variable. I'm sure

Hi, Is it possible to recreate "smoothed" data sets in R, by performing a PCA and then reconstructing a data set from say the first 2/3 EOFs? I've had a look in the help pages and don't seem to find anything relevant. Thanks in advance, Laura Laura Quinn Institute of Atmospheric Science School of Earth and Environment University of Leeds Leeds LS2 9JT tel: +44 113 343 1596

Hi, I was wondering if anyone could tell me why prcomp() will "Invent" modes of variation in a PCA on identical replicates of data? I would have expected 50 (or whatever number) of identical replicates to return a null score in such an analysis (or at the least, all variables would share the same PC score). This is not the case and I was wondering could someone point me in the direction

Hi, Is there a neat way of working out the slope of a flat surface in R? Given (x,y,z) co-ordinates of the four corners of a square, is there a function which will allow me to calculate the "mean" slope of the surface in a given direction? Thanks in advance.. Laura Laura Quinn Institute of Atmospheric Science School of Earth and Environment University of Leeds Leeds LS2 9JT tel: +44

Hi all, I think I may be confused by different people/programs using the word rotation differently. Does prcomp not perform rotations by default? If I understand it correctly retx=TRUE returns ordinated data, that I can plot for individual samples (prcomp()$x: which is the scaled and centered (rotated?) data multiplied by loadings). What does it mean that the data is rotated from the

Hi, Just a quick query - if I'm creating a function to produce a number of histograms per page of output (one per column from a matrix), how can I pass the column name of the matrix into the title (or indeed to form part of the x-axis label)? TIA, Laura Laura Quinn Institute of Atmospheric Science School of Earth and Environment University of Leeds Leeds LS2 9JT tel: +44 113 343 1596 fax:

I am wanting to plot pca loadings onto sites superimposed on a contour map. There are a maximum of 20 sites at which loadings might appear - however due to the nature of my data, missing data has meant that some stations have not been included in some of the pca. For example, I am performing pca for 5 different variables at 20 different sites for a number of different time intervals and want to

Hi, I was wondering if anyone might be able to help. I am trying to run R on a remote machine, part of the model run I am attempting writes an external file output as a png (about 48 iterations per model run). I am running R 1.9.1 on SuSe9.0, and am accessing this via ssh from a Debian machine. Initially I used the command ssh -X IP.address and whilst I was able to run the model successfully the

I'm wanting to create a series of near-identical matrices via the addition of "white noise" to my starting matrix. Is there a function within R which will allow me to do this? Thank you Laura Quinn Institute of Atmospheric Science School of Earth and Environment University of Leeds Leeds LS2 9JT tel: +44 113 343 1596 fax: +44 113 343 6716 mail: laura at env.leeds.ac.uk

I have a list containing 48 objects (each with 30 rows and 4 columns, all numeric), and wish to produce 4 boxplot series (with 48 plots in each) , one for each column of each object. Basically I want a boxplot from boxplot(mylist[[]][,i]) for i in 1:4. It seems that I can create a boxplot of length 48 from the entire list, but I don't seem able to subscript to return 4 boxplots from the list

Ok, I have reinstalled R-1.9.0 and this appears to have fixed the problems I was having with png(). However, I have a further question regarding png() Is it possible to pass a par() argument to the png() command? I am wanting to produce 4 plots per object, which I normally acheive on an X window by par(mfrow=c(1,4)). I have tried calling a new plot and setting par in this way but this has no

Apologies for the mass mailing today! I am attempting to produce a contour plot for phsical data on a map matrix. I have a small number of data points which each has an (x,y) co-ordinate together with a corresponding value which I would like to cvreate a contour plot for. I have tried the following code: contour(data$x,data$y,data$value) but am told: Error in contour.default(data$x, data$y,

Hello, Is there a simple way of stacking/merging two dataframes in R? I want to stack them piece-wise, not simply add one whole dataframe to the bottom of the other. I want to create as follows: x.frame: aX1 bX1 cX1 ... zX1 aX2 bX2 cX2 ... zX2 ... ... ... ... ... aX99 bX99 cX99 ... zX99 y.frame: aY1 bY1 cY1 ... zY1 aY2 bY2 cY2 ... zY2 ... ... ... ... ... aY99 bY99 cY99 ...

I have fitted an arima(0,0,2) model to my data, and am trying to plot a forecast for the next 15 time steps, but each time i try I am given the following error message: Error in .cbind.ts(list(...), makeNames(...), dframe = dframe, union = TRUE) : non-time series not of the correct length I am calculating as follows: my.arima<-arima(my.list,order=c(0,0,2))

Hi all, I am attempting to run PCA on a matrix (nrow=66, ncol=84) using 'prcomp' (stats package). My data (referred to as 'Q' in the code below) are separate river streamflow gaging stations (columns) and peak instantaneous discharge (rows). I am attempting to use PCA to identify regions of that vary together. I am entering the following command:

Hi, I was wondering if there is a way of plotting smoothed contours of PCA loadings onto a map. At the moment I have the loadings plotted as positive/negative (red/blue) scaled circles centred around where the physical stations are on a map - but I was wondering if there is a nicer way of plotting this data, perhaps as contours of "regions of influence" or somesuch? Any suggestions

I have a total mental block and can't find my way around this seemingly simple problem: I have created a function such that: my_answer_1=myfuntion(my_input_1) I am wanting to perform this calculation over a large number of datasets, but am having real difficulty calling and assigning - i think the problem les in the fact that I need to paste for call and assign. this is my best attempt so