similar to: step.lm() fails to drop {many empty 2-way factor cells} (PR#3527)

Exec. Summary: step() basically ``fails'' whereas MASS' stepAIC() does work This may not be a bug in the strictest sense, but at least something for the wish list. Unfortunately I have no time currently to investigate further myself but want to be sure this won't be forgotten: The example is using a real data set with 216 observations on 9 variables -- where we have

Folks, There appears to be a bug(?) in step() when used to screen logistic models. The problem appears to be specific to 1.2.1 (or maybe also 1.2.0?), as the proper behavior was observed in earlier versions. When I did the same on the surrogate log linear model, everything seemed okey. The data involved was the detergent data found in earlier editions of MASS, as given below, > detg1

Hi, I'm a new member here in the list. I am a graduate from University of Georgia. Recently in doing analysis using lme on a dataset, I found several questions: 1. How to express the equation when the correlation structure is very complicated. For exmaple, if the fixed is y(t)=0.03x1(t)+1.5x2(t)(I omitted "hat" and others). And the model with corARMA(p=2,q=3) is proper. What will be

Dear all, I have run the confint() function according to below and I get the following error message: > confint(stepAIC.glm.spe.var.konn2.abund, level=0.95) Waiting for profiling to be done... Error: missing value where logical needed In addition: Warning message: NaNs produced in: sqrt((fm$deviance - OriginalDeviance)/DispersionParameter) or > confint(stepAIC.glm.spe.var.konn2.abund,

I am hoping to get some advise on the following: I am looking for an automatic variable selection procedure to reduce the number of potential predictor variables (~ 50) in a multiple regression model. I would be interested to use the forward stepwise regression using the partial F test. I have looked into possible R-functions but could not find this particular approach. There is a function

Hello, I have some trouble using step() and stepAIC() functions. I'm predicting recruitment against several factors for different plant species using a negative binomial glm. Sometimes, summary(step(model)) or summary(stepAIC(model) does not select the best model (lowest AIC) but just stops before. For some species, step() works and stepAIC don't and in others, it's the opposite.

Hi all, I've experienced this problem using step() and stepAIC() when a name of a regressors has blanks in between (R:R1.7.0, os: w2ksp4). Please look at the following code: "x" <- c(14.122739306734, 14.4831100207131, 14.5556459667089, 14.5777151911177, 14.5285815352327, 14.0217803203846, 14.0732571632964, 14.7801310180502, 14.7839362960477, 14.7862217992577)

oops, Forgot to cc to the list. Regards, Mike -----Original Message----- From: dr mike [mailto:dr.mike at ntlworld.com] Sent: 24 February 2005 19:21 To: 'Spencer Graves' Subject: RE: [R] Bayesian stepwise (was: Forward Stepwise regression based onpartial F test) Spencer, Obviously the problem is one of supersaturation. In view of that, are you aware of the following? A Two-Stage

Dear All, For logistic regression models: is it possible to use validate (rms package) to compute bias-corrected AUC, but have variable selection with AIC use step (or stepAIC, from MASS), instead of fastbw? More details: I've been using the validate function (in the rms package, by Frank Harrell) to obtain, among other things, bootstrap bias-corrected estimates of the AUC, when variable

dear R users, I have fit the lm() on a mtrix of responses. i.e M1 = lm(cbind(R1,R2)~ X+Y+0). When i use summary(M1), it shows details for R1 and R2 separately. Now i want to use stepAIC on these models. But when i use stepAIC(M1) an error message comes saying that dropterm.mlm is not implemented. What is the way out to use stepAIC in such cases. regards,

Dear all, I cannot seem to get the R functions step or stepAIC to perform forward or stepwise regression as I expect. I have enclosed the example data in a dataframe at the end of this mail. Note rubbish is and rnorm(17) variable which I have deliberately added to the data to test the stepwise procedure. I have used wateruse.lm<-lm(waterusage~.,data=wateruse) # Fit full model

The birth weight example from ?stepAIC in package MASS runs well as indeed it should. However when I change stepAIC() calls to step() calls I get warning messages that I don't understand, although the output is similar. Warning messages: 1: In model.response(m, "numeric") : using type="numeric" with a factor response will be ignored (and three more the same.) Checked

Hello, I have a question pertaining to the stepwise regression which I am trying to perform. I have a data set in which I have 14 predictor variables accompanying my response variable. I am not sure what the difference is between the function "mle.stepwise" found in the wle package and the functions "step" or "stepAIC"? When would one use

Hi, First of all, thanks for the great work on R in general, and MASS in particular. It's been a life saver for me many times. However, I think I've discovered a bug. It seems that, when I use weights during an initial least-squares regression fit, and later try to add terms using stepAIC(), it uses the weights when looking to remove terms, but not when looking to add them:

Dear all, I have a model with simple terms, quadratic effects, and interactions. I am wondering what to do when a variable is involved in a significant interaction and in a non-significant quadratic effect. Here is an example d = data.frame(a=runif(20), b=runif(20)) d$y = d$a + d$b^2 So I create both an simple effect of a and a quadratic effect of b. m = lm(y ~ a + b + I(a^2) + I(b^2) +

Dear all, I'm attaching a patch that should fix the issue mentioned above. It simply makes the same check seen in the same file for global variables: emitPTXAddressSpace(PTy->getAddressSpace(), O); if (GVar->getAlignment() == 0) O << " .align " << (int) TD->getPrefTypeAlignment(ETy); else O << " .align " <<

I'm trying to build a SEM using the sem package. I'll attach my model specification below. When I turn debug=TRUE, it seems as if I'm getting to convergence because I get this message: Successive iterates within tolerance. Current iterate is probably solution. However, at the end of the process I get this message: Warning message: In sem.default(ram = ram, S = S, N = N,

Dear useRs, I need to compute zero of polynomial function fitted by lm. For example if I fit cubic equation by fit=lm(y~x+I(x^2)+i(x^3)) I can do it simply by polyroot(fit$coefficients). But, if I fit polynomial of higher order and optimize it by stepAIC, I get of course some coefficients removed. Then, if i have model y ~ I(x^2) + I(x^4) i cannot call polyroot in such way, because there is

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 06/16/2011 01:47 PM, Rob James wrote: > Ben, > > Thanks for this. Very helpful and clearly others have tripped over the > same problem > I would have supposed that the solution was to ask lrm (or glm) to use > LR rather than Wald, but I don't see syntax to achieve this. Typically drop1 or dropterm (MASS package) will drop

Thank you for Prof.Ripley's suggestion. I fixed the program by adding a lower scope, and the program ran, but I still got warning messages, and don't know what is going on, would this affect my results? ... Step: AIC= 12337.74 Surv(tlfup, cen) ~ MI[[j]]$trt + MI[[j]]$agem40 + MI[[j]]$agem40sq + mhtypeed1 + mhtypeed2 Df AIC <none> 12338 -