[R] [R-pkgs] The *makesweave* Package for Using Make with Sweave Efficiently
Hi R Community, I've just completed the initial version of (Linux-based) R package, called *makesweave*, for using Make to build Sweave documents efficiently. The idea is that R is started once (per shell) as a background process, then Make builds each Sweave source file, using the same R instance. More information in the package vignette. Feedback very welcome. regards -- Charlotte Maia http://sites.google.com/site/maiagx ___ R-packages mailing list r-packa...@r-project.org https://stat.ethz.ch/mailman/listinfo/r-packages __ R-help@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.
[R] [R-pkgs] Yet Another Package for Time Data
Hi fellow R developers/users, I've recently revised a package called rtv, and now consider it reasonably stable. Description: A package for conveniently representing, manipulating and visualising time data. Here, time is regarded as a random variable, and objects are used to represent realisations of that random variable. This is particularly useful for change points, irregular timeseries and failure events. There's a strong emphasis on continuous representations of time, with user-specified origins and units. http://cran.r-project.org/web/packages/rtv/index.html The package contains classes notably similar to the POSIXt classes. However, the classes are based on a somewhat different philosophy. The major advantage of the package, is that crtv objects (similar to POSIXct objects) allow user-specified origins and units. It also allows time events to be represented as fractional months or fractional years. More info in the package vignette. Bug reports and feature requests welcome. kind regards -- Charlotte Maia http://sites.google.com/site/maiagx ___ R-packages mailing list r-packa...@r-project.org https://stat.ethz.ch/mailman/listinfo/r-packages __ R-help@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.
[R] The KJV
Hey all, Does anyone know if there are any R packages with a copy of the KJV? I'm guessing the answer is no... So the next question, and the more important one is: Does anyone think it would be useful (e.g. for text-mining purposes)? I know almost nothing about theology, so I'm not sure what kind of questions theologists might have (that R could answer). An alternative, that would achieve a similar result (I think), would be an R interface to another open source system, such as Sword. kind regards -- Charlotte Maia http://sites.google.com/site/maiagx __ R-help@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.
[R] Emacs vs Eclipse vs Rcmdr
Hi everyone, I could have posted this on R-devel or the GUI list, however don't feel it's that serious. Hence, decided R-help would be the most appropriate. I'm not so much interested in which is the best user interface for R. Rather which is the best ***platform*** for developing ***new*** user interfaces for R. Noting I'm using the term user interface is a very general sense. (i.e. Can include anything from console/pseudoterminal widgets, to text editors with customised syntax highlighting, to elaborate menus and dialog boxes). Here are my initial thoughts: Emacs Pros: - A lot of computer "experts" use it. - Plus some high profile R people are involved in the development of ESS. - High level of customisation. Emacs Cons: - Need to know Lisp. - Counter intuitive. - It's really ugly. - No decent widget set (which is probably why it's ugly). Eclipse Pros: - It's kind of fashionable and nice looking. Eclipse Cons: - Unnecessarily complicated. - Need to know SWT (and maybe XML too?). - The process for installing (and finding) add on packages, is terrible. Rcmdr Pros and Cons: - I haven't used it for a long time, so can't really comment. - However, I was surprised by how many reverse dependencies it has. So I will assume it has some potential. Other people's thoughts welcome... kind regards -- Charlotte Maia Open Source Developer and Statistician http://sites.google.com/site/maiagx __ R-help@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.
[R] [R-pkgs] Multivariate ECDFs
Hey R people, I have just put a package on CRAN, mecdf 0.2.1. It computes multivariate ECDFs. i.e. Estimates (or perhaps I should say evaluates) a multivariate cumulative distribution function, using data, without any assumptions per se. Plus contrary to my own advise, the vignette contains some pretty pictures of the bivariate normal... The current package is relatively simple. However, I have searched for this topic, and found very little information, so I regard this as a new(ish) topic. The package is still new and relatively untested. I'm planning to implement smooth(ish) distribution functions in the near future. e.g. Interpolating the step function (possibly with some other adjustments). Suggestions (including criticisms) welcome, especially while the package is new. Merry Christmas, group hug... -- Charlotte Maia http://sites.google.com/site/maiagx/home ___ R-packages mailing list r-packa...@r-project.org https://stat.ethz.ch/mailman/listinfo/r-packages __ R-help@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.
[R] Re shorten str() output for long list
I doubt str is intended to work the way you want it to. I certainly wouldn't use it that way. The choice of data structure here is inappropriate, use a vector not a list. If you absolutely must do what you are asking, then simply write your own function. Use a heuristic for object length, say if less than 200 do such and such, otherwise do something else. In regards to some of your other posts, I think the R documentation is very good. Have you ever tried writing software documentation? It's a lot of work, which makes the R documentation in the standard packages quite impressive. If using a function for a subject area I'm not familiar with, I expect the unexpected. For mature software, the arguments section is extremely useful. Often when I'm using a function I forget the exact arguments, so have to look them up. For functions that I'm not familiar with, or unclear about, then I read the details section. Sometimes I might read it several times. -- Charlotte Maia http://sites.google.com/site/maiagx/home (this is a response to the website, not the MIME digest) __ R-help@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.
[R] Re Off topic - Compendium of distributions
Hi, I am going to sound mean here, however I don't feel the document is "very comprehensive". Maybe concise is a better word. I quickly looked through the document. The biggest problem is that there is very little discussion on multivariate distributions. Noting that multivariate distributions play a critical role in statistical theory, plus are gaining an increasing number of applications in various fields. I saw no discussion of distributions with circular, cylindrical, or spherical sample spaces (and for that matter the descriptions of bounded and unbounded are wrong). No discussion of copulas and no discussion of distributions that mix discrete and continuous random variables. Plus despite saying that your organisation supports nonparametric approaches, there was no obvious serious examples (maybe they are hidden in there somewhere...). It's important to note that as this is an R mailing list, many readers will have degrees in computer science, mathematics and statistics. Plus many will have looked at CRANs task view for distributions. I'd be surprised if there are many readers who are not already familiar with basic distributions such as normal, univariate, etc. Perhaps for people that have done one, maybe two, statistics courses at university, and majored in something entirely different, they very well may be quite impressed. Whilst this post is critical, I genuinely wish you the best of luck with your business, I hope your stakeholders and clientele are impressed by it. regards -- Charlotte Maia http://sites.google.com/site/maiagx/home (This is a response to website, not the MIME digest). __ R-help@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.
Re: [R] Replying to Posts Within Same Thread
On Thu, Dec 3, 2009 at 1:15 AM, Ted Harding wrote: > There is an issue which does not seem to have been raised yet in > this thread. In her original post, Charlotte Maia said she was > receiving messages in Digest form. > > There are two Digest options: "Get MIME or Plain Text Digests?" > You can choose which one to use, if you opt for Digest format, > when setting your subscription options in the R-help configuration > page at > > https://stat.ethz.ch/mailman/options/r-help > > In a plain-text Digest, all the messages are bundled into a single > message. Therefore when you reply to one of these you have to edit > out everything except the bit you are replying to; and also the > "Thread-relevant" headers of the original messages will heve been > suppressed -- only headers for the Dugest message will be present, > and these will have nothing to do with the threads that the separate > original messages belonged to. > > On the other hand, in a MIME Digest, each of the original messages > is attached as a separate attachment, along with its original headers. > A mail agent which is "MIME-Digest-aware" will allow the user to > open up each attachment separately, as a single message, and reply > to that. Then the mail agent *should* incorporate these headers, > including those which identify the thread. > > So, if you use Digest format, choose MIME Digest, and use a mail > agent which works properly! > > Ted. Hi, I have changed my settings from plain text to MIME (doesn't apply to this post, which is a response to a regular email, with r-help on the cc line). We'll see how things go, if my future posts appear threaded properly then we can assume this was the issue. If not, then I'll keep playing around. Me being the old-school kind of computer geek that I am, naturally go for plain text... thanks to all those who responded to this topic -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
[R] Again on overlaying plots (a plot region within a plot region)
Hi Ottorino-Luca, I'm not sure how flexible the layout command is off the top of my head. I have never seen an example of what you describe using layout, so I will assume it can't be done. One option is to use grid and lattice. This is possibly the best option if you are planning to do a lot of this sort of thing. Another option is to plot the second plot, on top of the first plot, using primitives such as points and lines, transforming your coordinates so that they match the coordinates of the first plot. This will be very difficult if you wish to put titles and axis on the nested plot, however simple if you restrict yourself to points and lines. Finding documentation and examples of R's graphical primitives should be easy. At the risk of sounding crude, I'm pretty sure both latex and many word processors allow one position one figure on top of another. This would seem the simplest solution for a one off job. Further to my post on R-devel, I recommend not using png format, except where absolutely necessary. regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
[R] Re How to remove R banner?
Hi, Annoying I love it (except possibly when re-directing standard out to a file). I think it's one of the command line options (in which case, it will be clearly documented). No idea how it works on the GUI systems (maybe an option somewhere...). In saying that, I have to assume that if people (you?) want control over R, they are using the *command line in the first place. The GUI systems represent a complete loss of control (esp the mac version, omg). There's always that in-between group who use Emacs/ESS, won't go there... *I use gnome-terminal + bash, plus a mixture of text editors. regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
Re: [R] Replying to Posts Within Same Thread
On 12/2/09, John Sorkin wrote: > I don't know what you are doing wrong because I don't know exactly what you > are doing. I do know that I don't have your problem when I simply reply to a > message without touching the subject line. > John Could you clarify the notion of "simply reply". There is no obvious reply links in either the archive page, or the digest. In the case, where someone sends me an email, such have you have done, I can simply reply. However, in general this option is not available. -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
[R] Replying to Posts Within Same Thread
Hi, When I reply to a post, it generally appears as a separate thread, rather than branching off the original post. Anyone know what I'm doing wrong? Noting I am currently receiving a couple of the lists in digest form. kind regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
Re: [R] Distance between sets of points in transformed environmental space
Hi Corrado, I was thinking about this some more. Maybe you could use a linear discriminate, i.e. a (hyper)plane that partitions your points into two sets, such that the misclassification rate is minimised. Closeness could be regarded as the number of misclassified points. Two sets would be distant, if no points are misclassified. I am assuming there is a standard function in R to do this, no idea what it is though. Plus this is a reasonably well known technique. Again the size of the sets needs to be accounted for. As well as the question, does the distance of set A from B, need to be the same as the distance of set B from A. Both the nearest neighbour approach and the discriminant approach, don't necessarily satisfy this condition. regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
Re: [R] Distance between sets of points in transformed environmental space
Well, here's another naive post from me (hopefully better than the last one). Firstly I'm not sure computing euclidean distance is that simple. I would assume temperatures and precipitation would need to be standardised in some way. I think the notion of how far away something is, and how distinct location wise something is, are quite different, so maybe two measures? For distance per se, I think your first idea is the best. Plus simple is always good... For distinctness, given one one of two sets, for each point, you could just compute the closest point to it. If the closest point is a member of the same set, we will call that a + point, if the closest point is a member of the other set, we will call it a - point. In principle the measure of distinctness would be the sum of the +'s, however there might need to be some scaling to take into account the number of points in each set. There are also a lot of fancy things out there, so someone will probably come up with a much fancier (and possibly better) idea than this. Well, that's just my rant, before I go to bed. kind regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
Re: [R] Histogram and Density... Sorry for Mispost
Hi, I just responded this post, then have just realised it was already responded to last month with a similar answer. I'm very sorry, I just started reading this mailing list a few days ago. Should be more aware of the discussion in future. -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
[R] R] Re: Histogram and Density on the the same graph
Hi, The computer I'm using now doesn't have R on it, so this is a naive guess: lines (density (y) ) regards -- Charlotte Maia http://sites.google.com/site/maiagx/home __ R-help@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.
[R] Backfitting with Missing Explanatory Values
Hi, I just wanted to check I'm not re-inventing the wheel here. I'm developing a new algorithm for backfitting (i.e. additive models) and for computing partial residuals, where partial residuals are still computed even where there are missing values. Noting additive models here contain both linear terms and smooth terms. If I am re-inventing the wheel could some one please let me know. I'm kind of on my own at the moment, and don't have quite as much academic support as I would like. Here's an excerpt from my incomplete package (on cran), amba. One way to think of residuals, is as some vector of values. If we start with the response values and subtract the overall mean, we get values with relatively high variance. If we then subtract the fitted values for the first term, the variance decreases. If we repeat for each term, the variance gradually decreases, until we are left with values with relatively low variance. In the ideal case, the residuals would have zero variance. If we apply certain special conditions, then it is possible to only subtract a fitted value, where the corresponding explanatory value is valid (i.e. not missing). Where it is not valid, we just skip that subtraction operation (i.e. for that particular observation, the variance is not reduced as much). For this to work, each explanatory variable's partial residuals for each fit (not just the final fit) must be zero-centered. For smoothers this isn't a big issue, however conventional linear terms often do not satisfy this zero-centered condition. Noting the centering condition applies to partial residuals in relation to an explanatory variable (not in relation to a parameter) and each explanatory may have multiple parameters associated with it. For our linear terms to satisfy it, we require extra parameters. Categorical terms require one parameter for each level, and polynomial terms, their own intercepts. kind regards Charlotte __ R-help@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.
