similar to: using poly in a linear regression in the presence of NA f ails (despite subsetting them out)

I ran into a to me surprising result on running lm with an orthogonal polynomial among the predictors. The lm command resulted in Error in qr(X) : NA/NaN/Inf in foreign function call (arg 1) Error during wrapup: despite my using a "subset" in the call to get rid of NA's. poly is apparently evaluated before any NA's are subsetted out of the data. Example code (attached to

Dear all, I am using poly() in lm() in the following form. 1> DelsDPWOS.lm3 <- lm(DelsPDWOS[,1] ~ poly(DelsPDWOS[,4],3)) 2> DelsDPWOS.I.lm3 <- lm(DelsPDWOS[,1] ~ poly(I(DelsPDWOS[,4]),3)) 3> DelsDPWOS.2.lm3 <- lm(DelsPDWOS[,1]~DelsPDWOS[,4]+I(DelsPDWOS[,4]^2)+I(DelsPDWOS[,4]^3)) 1 and 2 lead to identical but wrong results. 3 is correct. Surprisingly (to me) the residuals

I am having some trouble getting reftex, in particular the bibtex related features, to work properly in xemacs when editing text in Sweave files. I have added (defun Rnw-mode () (noweb-mode) (if (fboundp 'R-mode) (setq noweb-default-code-mode 'R-mode))) (add-to-list 'auto-mode-alist '("\\.Rnw\\'" . Rnw-mode)) (add-to-list 'auto-mode-alist

Full_Name: Russell Lenth Version: 2.6.2 OS: Windows XP Pro Submission from: (NULL) (128.255.132.36) The poly() function allows a higher-degree polynomial than it should, when raw=FALSE. For example, consider 5 distinct 'x' values, each repeated twice. we can fit a polynomial of degree 8: ===== R> x = rep(1:5, 2) R> y = rnorm(10) R> lm(y ~ poly(x, 8)) Call: lm(formula = y ~

Often in practical situations a predictor has missing values, so that poly crashes. For instance: > x<-1:10 > y<- x - 3 * x^2 + rnorm(10)/3 > x[3]<-NA > lm( y ~ poly(x,2) ) Error in poly(x, 2) : missing values are not allowed in 'poly' > > lm( y ~ poly(x,2) , subset=!is.na(x)) # This does not help?!? Error in poly(x, 2) : missing values are not allowed in

Hi, I am curious about how to interpret the results of a polynomial regression-- using poly(raw=TRUE) vs. poly(raw=FALSE). set.seed(123456) x <- rnorm(100) y <- jitter(1*x + 2*x^2 + 3*x^3 , 250) plot(y ~ x) l.poly <- lm(y ~ poly(x, 3)) l.poly.raw <- lm(y ~ poly(x, 3, raw=TRUE)) s <- seq(-3, 3, by=0.1) lines(s, predict(l.poly, data.frame(x=s)), col=1) lines(s,

By any chance is anyone aware of an R function that duplicates Octave's poly function? Here is a description of Octave's poly function: Function File: poly (A) If A is a square N-by-N matrix, `poly (A)' is the row vector of the coefficients of `det (z * eye (N) - a)', the characteristic polynomial of A. As an example we can use this to find the eigenvalues

How can I get the result of, e.g., poly(1:3. degree=2) to give me the unnormalized integer coefficients usually used to explain orthogonal polynomial contrasts, e.g, -1 1 0 -2 1 1 As I understand things, the columns of x^{1:degree} are first centered and then are normalized by 1/sqrt(col sum of squares), but I can't see how to relate this to what is returned by poly(). >

Bug in predict.lm & poly The predict function doesn't work when used with poly and newdata. For example, I'd expect the following code to work, and plot a fitted cubic to the nearly straight line: x <- 1:10 y <- x + rnorm(10)/100 plot(x,y) fit <- lm(y ~ poly(x,3)) newx <- seq(1,10,len=100) lines(newx,predict(fit,newdata=data.frame(x=newx))) However, the plotted

Hi all, I want to fit data called "metal" with a polynominal function as dP ~ a.0 + a.1 * U0 + a.2 * U0^2 + a.3 * U0^3 + a.4 * U0^4 The data set includes, the independant variable U0 and the dependant variable dP. I've seen that the combination of lm() and poly() can do that instead of using the nls() function. But I don't get how to interpret the results from the linear

I have a list of polygons generated by the contourLines() command (each object of the list is a list in itself with two objects: a vector of x values, and a vector of y values for each vertex). I wish to convert that list into a gpc.poly object of multiple contours. How do I do this? gpclib apparently has no method of coercing lists into the gpc.poly object type. As well, when I have a

Hello R-help I don’t know how to interpret significance from the contr.poly() function . From the example below : how can I tell if data has a significant Linear/quadratic/cubic trend? > contr.poly(4, c(1,2,4,8))               .L         .Q          .C [1,] -0.51287764  0.5296271 -0.45436947 [2,] -0.32637668 -0.1059254  0.79514657 [3,]  0.04662524 -0.7679594 -0.39757328 [4,]  0.79262909 

Full_Name: Jens Keienburg Version: 2.3.0 OS: Windows XP Submission from: (NULL) (193.174.53.122) I used the function lm() to calculate the coefficients of a polynome. If I used the function poly(t,2) to denote a polynome of form 1 + x + x^2, the coefficients are wrong. I appended an excerpt below: > t=1:100 > p=-20 - 10 * t + 2 * t^2 > p [1] -28 -32 -32 -28 -20 -8 8

When doing polynomial regression I believe it is a good idea to use the poly function to generate orthogonal polynomials. When doing this in Splus there is a handy function (transform.poly I think) to convert the coefficients produced by regression with the poly function back to the original scale. Has somebody written something similar for R ? Robert

Here's my little discussion example for a quadratic regression: http://pj.freefaculty.org/R/WorkingExamples/regression-quadratic-1.R Students press me to know the benefits of poly() over the more obvious regression formulas. I think I understand the theory on why poly() should be more numerically stable, but I'm having trouble writing down an example that proves the benefit of this. I

Dear list: since it took me a little while to figure out how to make use of naresid, I thought that the below R code might be useful as an example on the help page. Regards, Markus # generate some data x1 <- runif(20) y <- 10 + 5*x1 + rnorm(20) summary(lm.0 <- lm(y ~ x1)) # append some NA's to y y <- c(y, rep(NA, 5)) # generate some further x1s x1 <- c(x1, runif(5)) #

Dear list: since it took me a little while to figure out how to make use of naresid, I thought that the below R code might be useful as an example on the help page. Regards, Markus # generate some data x1 <- runif(20) y <- 10 + 5*x1 + rnorm(20) summary(lm.0 <- lm(y ~ x1)) # append some NA's to y y <- c(y, rep(NA, 5)) # generate some further x1s x1 <- c(x1, runif(5)) #

How do I go about generating a WEIGHTED mean (and standard error) of a variable (e.g., expenditures) for each level of a categorical variable (e.g., geographic region)? I'm looking for something comparable to PROC MEANS in SAS with both a class and weight statement. Thanks. Marc [[alternative HTML version deleted]]

Full_Name: David Clifford Version: Version 1.5.1 (2002-06-17) OS: Red Hat 7.3 Submission from: (NULL) (128.135.149.55) For n values above 100 there appears to be a bug in contr.poly(n). The contrast matrix should have rank n-1. Running the code below gives output (ie errors) at n=98, 100 and every value greater than 102. for(n in 2:150) { K <- contr.poly(n) rnk <-

Besides the primary citation, "Kennedy, W. J. Jr and Gentle, J. E. (1980) Statistical Computing Marcel Dekker." (which is $300 and my library doesn't have it), is there any other documentation on how to take a poly() object and predict "by hand" new data? E.g. What do those coefficients actually mean ("The orthogonal polynomial is summarized by the coefficients, which