[R] Transforming smooth functions of gam using mgcv

Hi, I have 2 questions, both pertaining to additive mixed models using mgcv(). I have fit several additive mixed models using the mgcv package (v 1.4) in R (v 2.8) modeling log chlorophyll as follows: mod<-try(gamm(log(Chl) ~ s(Year,bs="cs") + s(Latitude,Longitude,bs="tp") + s(Dayofyear,bs="c

Re: [R] glm problem

Thanks for the reply, that was a typo only in the post. as i was trying to remove some 'sensitive names'. In my real work, i made sure that 1) this command runs without problem qdata.glm = glm(X258 ~ ., family = binomial, data = qdata[ sample(dim(qdata)[1], 1000), ] ) 2) these two gave a warn

[R] anova help

Hi all, I am trying to run a two factor anova, but one of the factors is a random factor, now I am also running in SPSS and it seems its dividing by the wrong term to get the appropriate F term. here is my data. In SPSS the F scores about double the ones in R, how can I specify one of my factors as

[R] How to fit GARCH(1,1) with targeted unconditional variance?

Hello, I have a univariate data set, and the unconditional variance is 1. I would like to fit a GARCH(1,1) model to the data set with a constraint: \omega (the constant parameter in GARCH(1,1)) is equal to 1-\alpha-\beta. So the unconditional variance can be controled to be \omega /(1-\alpha-\

[R] extracting standard errors in glm.nb

Hi, How do I extract the standard errors of the coefficients in a glm.nb model? Thanks. -- View this message in context: http://www.nabble.com/extracting-standard-errors-in-glm.nb-tp22019202p22019202.html Sent from the R help mailing list archive at Nabble.com. ___

Re: [R] glm problem

Roger wrote: Thanks for the reply, that was a typo only in the post. as i was trying to remove some 'sensitive names'. In my real work, i made sure that 1) this command runs without problem qdata.glm = glm(X258 ~ ., family = binomial, data = qdata[ sample(dim(qdata)[1], 1000), ] ) 2) thes

Re: [R] Anova and unbalanced designs

Dear John - thank you for your detailed answer and help. Your answer encourages me to ask further: by choosing different contrasts, what are the different hypothesis which are being tested? (or put differently - should I prefer contr.sum over contr.poly or contr.helmert, or does this makes no diff

Re: [R] extracting standard errors in glm.nb

jpl math.unl.edu> writes: > > How do I extract the standard errors of the coefficients in a glm.nb model? library(MASS) quine.nb1 <- glm.nb(Days ~ Sex/(Age + Eth*Lrn), data = quine) sumnb1 = summary(quine.nb1) sumnb1 str(sumnb1) # tells you about the anatomy. Check for $coefficients Dieter __

Re: [R] anova help

Provide some example data & code and someone may be able to help. On Sat, Feb 14, 2009 at 2:46 PM, Joe King wrote: > Hi all, I am trying to run a two factor anova, but one of the factors is a > random factor, now I am also running in SPSS and it seems its dividing by > the wrong term to get the a

[R] How to plot image of contour in percent?

I have searched this forum with keywords contour and percent, it only give one irrelevent result. My problems is how to plot contour in percent. In my figure, I have use kde2d to generate density of an array with two dimension point. Using image and contour could plot it. But the line contour gener

[R] PRESS / RMSEP

Dear all , I want to do PRESS (prediction error sums of squares) or the residual mean square error of prediction (RMSEP) which will give me value that is valid for 'future predictions of independent data'. I am using different methods for example, Multiple Linear Regression, L

[R] Package for Markov (Regime) Switching (ARMA) Models

Hello R-Users Is there a package in R, that handles Markov (Regime) Switching (ARMA) Models for time series modelling and prediction? Thank you very much. Regards, Andreas. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listi

Re: [R] Website, book, paper, etc. that shows example plots of distributions?

On Feb 14, 2009, at 6:48 PM, Jason Rupert wrote: > Many thanks to Greg L. Snow and David Winsemius for their responses. > > First off I can safely say I don't know enough statistics to be > dangerous, but hopefully I will get to that point:) > > Regarding the goal - ultimately I would like to u

Re: [R] How to plot image of contour in percent?

On Feb 15, 2009, at 7:55 AM, yk wrote: I have searched this forum with keywords contour and percent, it only give one irrelevent result. My problems is how to plot contour in percent. In my figure, I have use kde2d to generate density of an array with two dimension point. Using image and contou

Re: [R] Anova and unbalanced designs

Dear Tal, A complete explanation of this issue is too long for an email; I do address it in my Applied Regression Analysis and Generalized Linear Models text. The question seems to come up so frequently that Georges Monette and I are writing a paper on it (and related issues) for the upcoming useR

Re: [R] extracting standard errors in glm.nb

Hi Dieter, I don't see anything in the str(model) output that matches the standard errors of the coefficients given in summary(model). Any other suggestion? Thanks. jpl Dieter Menne wrote: > > jpl math.unl.edu> writes: > >> >> How do I extract the standard errors of the coefficients in

[R] Kalman Filter - dlm package

Dear all, I am currently trying to use the "dlm" package for Kalman filtering. My model is very simple: Y_t = F'_t Theta_t + v_t Theta_t = G_t Theta_t-1 + w_t v_t ~ N(0,V_t) = N(0,V) w_t ~ N(0,W_t) = N(0,W) Y_ t is a univariate time series (1x1) F_t is a vector of factor retu

Re: [R] extracting standard errors in glm.nb

Look more thoroughly? You should see this about 2/3 the way down the output: $ coefficients : num [1:14, 1:4] 3.019 -0.475 -0.709 -0.724 -0.615 ... ..- attr(*, "dimnames")=List of 2 .. ..$ : chr [1:14] "(Intercept)" "SexM" "SexF:AgeF1" "SexM:AgeF1" ... .. ..$ : chr [1:4] "Estimate" "St

Re: [R] OT: A test with dependent samples.

Dear List, Catching up with my backlog, I stumbled upon this : On Wed, 11 Feb 2009 10:33:13 +1300, Rolf Turner wrote : > I am appealing to the general collective wisdom of this list in respect > of a statistics (rather than R) question. This question comes to me > from a friend who is a veterin

[R] loading mgcv package

Hi ,When I try to load the 'mgcv" package, often, but not always, get this error message. What am I doing wrong? I even tried reinstalling a few times. __ This is mgcv 1.4-1.1 Error in runif(1) : .Random.seed is not an integer vector but of type 'list' Error : .onAttach failed in

[R] Fast ave for sorted data?

Hi, This is probably really obvious, by I can't seem to find anything on it. Is there a fast version of ave for when the data is already sorted in terms of the factor, or if the breaks are already known? Basically, I have: X = 0.1, 0.2, 0.32, 0.32, 0.4, 0.56, 0.56, 0.7... Y = 223, 434, 343, 5

Re: [R] anova help

Hi Joe, you might want to have a look at the nlme package with the lme command. Another option is the more advanced lmer package. Lastly, you could have a look at the ?aov command, and notice the option of using the +Error() term (but that would only work for balanced design cases, so I've heard,

[R] MDS with ranking data (and transformation)

Dear Sirs and madams :-) I am trying to teach myself multidimensional scaling. To that effect I have collected a survey asking people to rank 10 philosophers and politicians according to their preference. I have collected 61 answers. The data is organized in ten columns and 61 rows. the columns a

[R] How to add new corStruct class in lme?

I was trying to fit a linear mixed model with a user specified correlation structure with lme (package nlme). In the book of Pinheiro and Bates (2000), it was mentioned that new corStruct classes can be added, and the corAR1 constructor and methods can serve as templates. But I am just too dumb t

Re: [R] adding S3 vs. S4 to a new package

On 13/02/2009, at 5:42 PM, Edna Bell wrote: Dear R Gurus: I'm building a package and am wondering whether to use S3 or S4 methods. Any advice, please? Thanks, See fortune(121)! cheers, Rolf Turner

[R] Porcentual separation

Dear Fellows: Thanks you all for your great help. Without your guidance I'd not have been able to get this far. So this is the current problem: I have a series (dataframe with 1 column and 283 rows, all with numbers (no NA's) that go from 0 to 3 or 4 with up to 14 decimals. They indicate a

[R] mnormt and mvtnorm

Hi R user, I came across a question when I played around with mnormt package (mvtnorm as well), could any of you let me know where I made the mistake? Thanks! For instance, with a multivariate normal distribution with known mean and variance (to be simple, I tried the i.i.d case first) u1 <-

Re: [R] Fast ave for sorted data?

On Sun, 15 Feb 2009, Zhou Fang wrote: Hi, This is probably really obvious, by I can't seem to find anything on it. Is there a fast version of ave for when the data is already sorted in terms of the factor, or if the breaks are already known? If all you want are means, you can use rle() an

Re: [R] MDS with ranking data (and transformation)

On Sun, 15 Feb 2009, achristoffersen wrote: Dear Sirs and madams :-) I am trying to teach myself multidimensional scaling. To that effect I have collected a survey asking people to rank 10 philosophers and politicians according to their preference. I have collected 61 answers. The data is orga

Re: [R] Fast ave for sorted data?

Thanks! That does exactly what I want. (Heck, maybe this should be included as a default sorted alternative to ave.) I was thinking of doing it another way using cumsums, but maybe this method is faster. Zhou __ R-help@r-project.org mailing list https:

[R] Plot of MaximumLikihoodFunction

Hey I'm a relative newbie to R my question is: is it possible to plot the MaximumLiklihood function of the parameter estimation of a garch(arch,arma...) model?? Best regards Rainer __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/li

Re: [R] anova help

Ok well heres my data, with the second variable (B variable) being the random variable and the other fixed. Although the F values are about half of what SPSS puts out. My code is anova(lm(dependentrandom~typemusic+typemusic*musicselection)) This is just dummy data for a class but I am tryin

Re: [R] Overloading in R

Hi Shrikanth -- SHRIKANTH SHANKAR writes: > I have been trying to write a new class that reimplements the vector > class. As a test of my overloading I decided to try and and call > t.test on two vectors which were objects of my class rather than the > default class. > > The overloaded length I

[R] matrix transpose

This is supposed to be an easy operation, but R 2.8.1 on Mac OS X gives me a lot trouble. the t() function simply does not work properly. What I mean is it works sometimes but does not work at the most of the time, even with the same matrix. this is an example taken from R help > a <- matrix(1:3

Re: [R] matrix transpose

On 16/02/2009, at 8:36 AM, Roger wrote: This is supposed to be an easy operation, but R 2.8.1 on Mac OS X gives me a lot trouble. the t() function simply does not work properly. What I mean is it works sometimes but does not work at the most of the time, even with the same matrix. this

Re: [R] anova help

You probably did not forget to attach the data the first time, either. You, instead, probably did not remember to read the posting guide thoroughly regarding what constitutes an acceptable attachment type. Suggest that you change your mail client settings to transmit plain text and "follow

[R] Passing environments: why doesn't this work???

This is a trivial example I set up to see if I could pass an environment and use the variables in it (this is for a function that will be called many times and might need to use a lot of variables that won't be changing, so it seemed more sensible to use an environment). Here's the code: #

Re: [R] Passing environments: why doesn't this work???

John Tillinghast wrote: This is a trivial example I set up to see if I could pass an environment and use the variables in it (this is for a function that will be called many times and might need to use a lot of variables that won't be changing, so it seemed more sensible to use an environment).

[R] (no subject)

Dear all,I had just started to learn R. So my question may sound a bit stupid for you. Is that possible to match and merge two tables in R? I mean I have two tables, in the first I have two columns: RING NUMBER, WEIGHT and CAPTURE DATE, in the second RING NUMBER, SEX and CAPTURE DATE. 1) First I

Re: [R] matrix transpose

Roger wrote: > This is supposed to be an easy operation, but R 2.8.1 on Mac OS X gives me a > lot trouble. the t() function simply does not work properly. What I mean is > it works sometimes but does not work at the most of the time, even with the > same matrix. > > this is an example taken from R

Re: [R] matrix transpose

Here's one way that seems to work: a <- matrix(1:30,5,6) # Create a target for the transpose b <- matrix(1:(nrow(a)*ncol(a)),ncol(a),nrow(a)) # populate b with the transpose of a for (i in 1:ncol(a)) { b[i,1:(nrow(a))] <- a[1:(nrow(a)),i]} # Check: Did it work? all.equal(b,t(a)) I'm sure there's

Re: [R] matrix transpose

Rolf Turner wrote: > > On 16/02/2009, at 8:36 AM, Roger wrote: > >> This is supposed to be an easy operation, but R 2.8.1 on Mac OS X >> gives me a >> lot trouble. the t() function simply does not work properly. What I >> mean is >> it works sometimes but does not work at the most of the time, eve

Re: [R] (no subject)

Vincze Orsolya wrote: Dear all,I had just started to learn R. So my question may sound a bit stupid for you. Is that possible to match and merge two tables in R? I mean I have two tables, in the first I have two columns: RING NUMBER, WEIGHT and CAPTURE Err, ... for large values of "two"? Or is

Re: [R] Passing environments: why doesn't this work???

Note that you can do things like this too: env$x with(env, x^2) Also its possible you are really getting into object oriented programming without realizing it with the your environments being the objects your functions being the methods. In that case check out the proto package which formalize

[R] GLMM, ML, PQL, lmer

Dear R community, I have two questions regarding fitting GLMM using maximum likelihood method. The first one arises from trying repeat an analysis in the Breslow and Clayton 1993 JASA paper. Model 3 of the epileptic dataset has two random effects, one subject specific, and one observation specific

Re: [R] MDS with ranking data (and transformation)

Charles C. Berry wrote: > > Something like > > dat <- t( apply( your.data, 1 , order ) ) > colnames(dat) <- paste( 'philosopher', 1:10, sep='_' ) > > HTH, > Hi Chuck - thx - but I have trouble geting further. I think I need "rank" not "sort" - but this still doesnet solve my conundrum. I ha

Re: [R] (no subject)

Also see the sqldf package which lets you do this using SQL within R. http://sqldf.googlecode.com On Sun, Feb 15, 2009 at 3:43 PM, Peter Dalgaard wrote: > Vincze Orsolya wrote: >> >> Dear all,I had just started to learn R. So my question may sound a bit >> stupid for you. >> Is that possible to m

Re: [R] GLMM, ML, PQL, lmer

Youyi Fong u.washington.edu> writes: > > Dear R community, > > I have two questions regarding fitting GLMM using maximum likelihood method. > The first one arises from trying repeat an analysis in the Breslow and > Clayton 1993 JASA paper. Model 3 of the epileptic dataset has two random > effec

[R] feature extraction on time series data

Hi, This is a practical question and I am sure there are many statisticians can give me a hand. I have 500 time series data (500 rows), each row contains 100 intervals, i.e., on each row, I have X1, X2, . X100. I am trying to reduce the dimension of this input because the data at the end of

[R] Unadulterated plot

To all, Apologies if this question has already been asked but I can't find anything. I can't seem to think of more specific search terms. I want to display/create a file of a pure plot with a specific height and width. I want to utilise every single pixel inside the axes. I do not want to displ

[R] std error of the covariance matrix of random effects in lme4

Are the standard errors of the covariance matrix elements for the random effects in lme4 (lmer) available. That is, R using lme4 function lmer gives me Random effects: Groups NameVariance Std.Dev. Corr id (Intercept) 0.6360.797 time0.1550.394

[R] solve a linear system with argument is a matrix

Hi everyone, I have a question to solve linear equations with argument is a matrix. For example, X is a q by p matrix satisfies F(X)=A and t(B)%*%X+t(X)%*%B=0, where A and B are known q by p matrices and 0 is p by p matrix with all elements are 0s. And F(.) is a known function of matrix, X. Now,

Re: [R] extracting standard errors in glm.nb

Mea culpa. Thanks! jpl David Winsemius wrote: > > Look more thoroughly? > > You should see this about 2/3 the way down the output: > > $ coefficients : num [1:14, 1:4] 3.019 -0.475 -0.709 -0.724 -0.615 ... >..- attr(*, "dimnames")=List of 2 >.. ..$ : chr [1:14] "(Intercept)" "SexM

[R] How do I Edit Source Code Files On The Mac?

Hello! I'm using Mac OS X 10.4.11 on a G4. I use the "R" command, straight from a terminal, because the "GUI" is slower, and I've never liked GUIs anyway. I tried to set up my environment to use the vi editor, but it always tries to save it in wierd things like this: "/private/tmp/RtmpWIhhuo/4

Re: [R] solve a linear system with argument is a matrix

Hi, This looks like a Lyapanov equation. It is used in control theory I think. There might be something if you search in those packages that deal with this. HTH, On Feb 16, 9:25 am, �|珊�x wrote: > Hi everyone, > > I have a question to solve linear equations with argument is a matrix. > For ex

Re: [R] How do I Edit Source Code Files On The Mac?

On Sun, 15 Feb 2009, greggal...@gmail.com wrote: Hello! I'm using Mac OS X 10.4.11 on a G4. I use the "R" command, straight from a terminal, because the "GUI" is slower, and I've never liked GUIs anyway. I tried to set up my environment to use the vi editor, but it always tries to save it in

Re: [R] How to plot image of contour in percent?

I was thinking of a more focused search including the terms contour and either r-help or r-project, but not one restricted to GoogleGroups, ,,, more along the lines of: http://www.google.com/search?hl=en&rlz=1B3GGGL_enUS278US278&q=r-help+quantile+contour&start=20&sa=N or... http://www.goog

Re: [R] solve a linear system with argument is a matrix

Depending on F, there may already be software to solve your system. If not, have you considered recasting the problem as minimizing the sum of squares of deviations from solutions? Then you could use 'optim' or 'nlminb' or ... . Spencer Graves andrew wrote: > Hi, > > This looks like a Lyapanov e

[R] Comparison of age categories using contrasts

Dear listers, I would like to compare the levels of a factor with 8 age categories (0,10] (10,20] (20,30] (30,40] (40,50] (50,60] (60,70] (70,90] (however, the factor has not been ordered yet). The default in glm is cont.treatment (for unordered factors) and that leads to compare each level t

[R] McFadden Pseudo R2 with positive Likelihood??

Hi I'm calculating McFaddens pseudo R2 (1- (Lmodel/L0)) for some models, and i ran into a model where the model of interest has a positive likelihood, while the null has a negative. This of course results in odd values. I searched the list, but find no help. Can anyone provide insight? Thanks K

Re: [R] I want axes that cross

On Fri, Feb 13, 2009 at 3:14 PM, Marc Schwartz wrote: > on 02/13/2009 02:19 PM Paul Johnson wrote: >> On Fri, Feb 13, 2009 at 1:51 PM, Marc Schwartz > > OK, so given all of the above, something like the following should work: > > set.seed(1233240) > x <- rnorm(100) > z <- gl(2,50) > y <- rnorm(100

Re: [R] Comparison of age categories using contrasts

Hi Patrick, >> The default in glm is cont.treatment (for unordered factors) and that >> leads to compare each level to the first one. I would rather prefer to >> compare the 2nd to the 1st, the 3rd to the 2nd, the 4th to the 3rd, >> etc... The functions ?C and ?contrasts allow you to set up yo

[R] Time series

Hello everyone. I am trying to plot data from a time series in R and have run across some problems. I was wondering if someone could help. I have data taken every 15 minutes of a list of rankings of 25 titles. I want to plot one of the article's titles's ranks as a time series. I've been trying if

[R] Transformation of Variables

I have a data frame in R where all the variables are character in nature. kindly let know how I can transform these character variable into say "numeric","Categorical","continuous" etc.Please provide me with the proper syntax. Thank you in advance. -- View this message in context: http://www.