similar to: Is R good for not-professional-statistician, un-mathematical clinical researchers?

I am trying to develop a prognostic model using logistic regression. I built a full , approximate models with the use of penalization - design package. Also, I tried Chi-square criteria, step-down techniques. Used BS for model validation. The main purpose is to develop a predictive model for future patient population. One of the strong predictor pertains to the study design and would not

Often it is useful to keep a "codebook" to document the contents of a dataset. (By "dataset" I mean a rectangular structure such as a dataframe.) The codebook has as many rows as the dataset has columns (variables, fields). The columns (fields) of the codebook may include: ? variable name ? type (character, factor, integer, etc) ? variable label

Consider all the text that one sees on the console during an R session. Is there a way, within R, to make all this text--both the "output" and the "messages"--automatically get copied to a single text file, in addition to seeing it on the console? If I remember to save the console to a file at the end of my R session, that does it. But (1) That requires pointing and

I used read.dta() to read in a Stata 9 dataset to R. The "Sex01" variable takes on two values in Stata: 0 and 1, and it is labeled "M" and "F" respectively, analogous to an R factor. Thus, read.dta reads it in as a factor. Now, I wanted to see what this variable *really* is, in R. For instance, sometimes R converts a 0/1 variable into a 1/2 variable when it considers

We are finding more and more clinical researchers interested in learning R and are starting to teach R to clinicians. We have put some teaching material on our site: http://biostat.mc.vanderbilt.edu -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University

Dear all, There have been a variety of discussions on the R list regarding the use of R in clinical trials. The following post from the STATA list provides an interesting opinion regarding why SAS remains so popular in this arena: http://www.stata.com/statalist/archive/2008-01/msg00098.html Regards, -Cody Hamilton

Greetings! (and apologies upfront for any duplication as a result of my cross-posting) It''s been five weeks now since I began working with RoR. My experience has been incredibly positive. So much so that I''m emboldened to undertake a project I''d previously concluded I no longer had the technical chops to even _start_. I''ve included a brief description of

I would like to use lattice graphics to plot multiple functions (or groups or subpopulations) on the same plot region, using different line types "lty" or colors "col" to distinguish the functions (or groups). In traditional graphics, this seems straightforward: First plot all the data using 'type="n"', and subsequently execute a series of "points"

Is there any documentation on what kind of SPSS file can and cannot be read by read.spss? Alternatively, how can one modify or "clean" an SPSS file to make it readable by read.spss? What properties must a *.sav file before read.spss can read it? The file in this example is 270KB, with 5 rows and 173 columns. I have no trouble reading larger files with read.spss, so it's not

When we call a lattice function such as xyplot, to what extent does the "data" designation cause the function to look inside the "data" for variables? In the examples below, the "subset" argument understands that "Variety" is a variable in the data. But the "scales" argument does not understand that "nitro" is a variable in the data.

Is there a way to specify the color of the main title, the subtitle, or the axis labels? I mean, for instance, something like title(main="cougar", col=2) For me, the above command produces the color black; that is, the "col" argument has no effect. I'm on a Windows 2000 machine with > version _ platform i386-pc-mingw32 arch i386 os mingw32 system

How would I create the following plot using lattice? symbols( combPsummary$pastRate, combPsummary$finRate, circles=sqrt(combPsummary$N) ) The idea is to plot finRate vs pastRate using circles whose areas are proportional to the number of people in each group. The following attempt does not really work: xyplot( finRate ~ pastRate , data=combPsummary , panel=function( x, y,

I want to put the "plus or minus" symbol into a character variable, so that this can be turned into a factor and be displayed in the "strip" of a faceted ggplot2 plot. A very nice solution, thanks to Professor Ripley's post of Nov 16, 2008; 3:13pm, visible at http://r.789695.n4.nabble.com/Symbols-to-use-in-text-td874239.html and subsequently

help.search("groupedData") says that it's part of the lme4 package, but it appears not to be there (details below). Is this because lme4 is new and (perhaps) still under development? > update.packages() trying URL `http://cran.r-project.org/bin/windows/contrib/1.9/PACKAGES' Content type `text/plain; charset=iso-8859-1' length 19113 bytes opened URL downloaded 18Kb >

Suppose we have some glm object such as: myglm <- glm( y ~ x, data=DAT) Is there an elegant way--or the "right way" within the R way of thinking--to obtain the names of the response variable, the predictor variables, and the dataset, as character strings? For instance, suppose the "right way" was to use the (currently fictitious) functions theresponse(), thepredictors(),

There are three DCs in my Windows 2003 AD domain, but I have noticed that only one of them is referenced in my krb.conf and krb5.conf. Should there be a reference to one or two of the other domain controllers? If the DC goes down, how will my Samba/Winbind servers authenticate? -- Eric Robinson Disclaimer - February 27, 2011 This email and any files

The purpose of this email is to (1) report an example where fisher.test returns p > 1 (2) ask if there is a reliable way to avoid p>1 with fisher.test. If one has designed one's code to return an error when it finds a "nonsensical" probability, of course a value of p>1 can cause havoc. Example: > junk<-data.frame(score=c(rep(0,14), rep(1,29), rep(2, 16))) >

Hi all, I am trying to convert a data set of physician death codings (each individual's cause of death is coded by multiple physicians) from long to wide format, but the "reshape" function doesn't seem to work because it requires a "time" variable to identify the sequence among the repeated observations within individuals. My data set has no order, and different

How does one implement a likelihood-ratio test, to test whether the variances of the random effects differ between two groups of subjects? Suppose your data consist of repeated measures on subjects belonging to two groups, say boys and girls, and you are fitting a linear mixed-effects model for the response as a function of time. The within-subject errors (residuals) have the same variance in

The print method for lme *prints out* the fitted correlation matrix for the random effects. Is there any way to get these values as an object in R? I have examined the components of the lme object (called "junk" in the example below) and the components of summary(junk) without finding these numbers. (How I did this: I dumped the entire lme object to a text file and then used egrep to