[R] multivariate nonparametric regression with e = 0
Hello all, I'm a relatively new user of R, having mostly used it only for plotting so far. I'm also not very familiar with regression methods, hence forgive my greenness on the topic. What I want to do in R is multivariate nonparametric regression, with a slight hitch. From my experimental data I have a multitude of samples whose values approximate a function `f' that is defined over a 5D space (i.e., f: R^5-R). The values of the collected samples, call these `y', approximate `f', but due to the process by which they are collected, they always over-estimate (i.e., y = f + e, e = 0). The distribution of the error `e' can likely be modelled using the positive half of the normal distribution. Naturally I'm trying to obtain a smooth and relatively faithful approximation of `f' using the collected samples `y'. What would be the most fruitful approach in R to doing this? Even suggestions on which package/function to use would be tremendously helpful, as I don't yet know what their strengths/weaknesses are. Also, I would consider parametric regression as well, but in the general case I don't think I can assume/guess for my data at what the appropriate parametric basis functions should be... -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac/ __ 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
[R] label *on the side* in conditional lattice plots?
I'm doing bwplot(x ~ y | z, ...) with lattice, but would like the z-labels to appear to the *side* of each bwplot, rather than on top... is this possible? The main reason is that my bwplots are horizontal, and as such can be nicely squished in the vertical axis (big space-saver when including in papers), but the labels get in the way, look too large in comparison, while I have all this empty space to the side of the plots which is being wasted instead... -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac __ 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
Re: [R] label *on the side* in conditional lattice plots?
On 8/31/05, Wiener, Matthew [EMAIL PROTECTED] wrote: I think you might be able to use the horizontal argument to lattice to rotate all your plots and squish them in the other dimension. (Though I don't know whether you consider that a good outcome ...) I take it you mean to switch the layout so that the conditioned bwplots are side by side, narrow but tall, with horizontal=TRUE, and then incorporate into the paper by first rotating 90 degrees. But is there a way to rotate the axis labels then? Having the z-label written sideways is probably OK, but having the x- and y-axis labels also sideways is just too much. -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac __ 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
Re: [R] label *on the side* in conditional lattice plots?
On 8/31/05, Maciej Kalisiak [EMAIL PROTECTED] wrote: I take it you mean to switch the layout so that the conditioned bwplots are side by side, narrow but tall, with horizontal=TRUE, and Oops, I obviously meant horizontal=FALSE... -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac __ 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
Re: [R] plot(type=h) equivalent in Lattice?
Ah, great, thanks. I've actually ran into it just before I saw your posting, purely by accident, when I happened to look at the source for panel.xyplot while looking for something else, and noticed the type argument, and in particular the handling of type='h'... I didn't realize plot(type='h'...) makes use of plot.xy(), otherwise the plot.xy(graphics)-xyplot(lattice) move would have been obvious... -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] plot(type=h) equivalent in Lattice?
I tend to prefer doing graphics in R using the lattice library. I'm porting some old scripts. Is there a nice way to get in lattice the equivalent of the plot(type='h'), which is the high-density lines/histogram plot in base graphics package? I tried doing this with barchart(), but with limited success... the fact that I have ~1000 datapoints/bars causes problems, in that R insists on rendering a tick/label for each bar, and as far as I can tell the plot doesn't come out as accurate, due to running together of the bars (at least in plot(type='h') you are guaranteed that each bar/line is exactly one pixel wide)... -- Maciej Kalisiak [EMAIL PROTECTED] http://www.dgp.toronto.edu/~mac __ 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