[R] How to compare two curve model

Hi all: I have two curve models: model1-nls(result ~ exp(b0 + b1*(time)), start = list(b0 = 0, b1 = 5),trace=TRUE,data=data1) model2-nls(result ~ exp(b0 + b1*(time)), start = list(b0 = 0, b1 = 5),trace=TRUE,data=data2) I wanna compare the two models to find out whether the difference

[R] How to get intersection of multiple vectors?

v1-c(a,b,c,d) v2-c(a,b,e) v3-c(a,f,g) I want to get the intersection of v1,v2,v3,ie a How can I do then? What I know is only for 2 vectors via intersect function,but don't know how to deal with multiple vectors. Many thanks [[alternative HTML version deleted]]

Re: [R] subset

Many thanks for your elaborated explaina. At 2012-01-13 11:34:51,R. Michael Weylandt michael.weyla...@gmail.com wrote: As Jorge noted, the fix is to use %in%: a fuller explanation of why `==` didn't work is that it implicitly used vector recycling: look at with(data, id == c(a, c))

Re: [R] subset

Thanks£¡ ÔÚ 2012-01-13 10:51:14£¬Jorge I Velez jorgeivanve...@gmail.com дµÀ£º Hi, Use %in% instead of ==. HTH, Jorge.- On Thu, Jan 12, 2012 at 9:36 PM, ÃÏÐÀ wrote: Hi all I have a question about subset function. dat id x1 x2 x3 1 a 1 11 111 2 b 2 22 222 3 c 3 33 333 4 d

[R] subset

Hi all I have a question about subset function. dat id x1 x2 x3 1 a 1 11 111 2 b 2 22 222 3 c 3 33 333 4 d 4 44 444 subset(dat,id==c(a,c)) id x1 x2 x3 1 a 1 11 111 subset(dat,id==c(a,d)) id x1 x2 x3 1 a 1 11 111 4 d 4 44 444 From the above, if I choose id=a,c, the

Re: [R] 64bit R under 32bit winxp

Thanks At 2012-01-11 16:55:32,Jeff Newmiller jdnew...@dcn.davis.ca.us wrote: You cannot install 64-bit R on 32-bit OS, but you can install a 32-bit R on a 64-bit OS, and you can later install 64-bit R as well. That is, installing 32-bit R does not interfere with your option to later install

[R] 64bit R under 32bit winxp

Hi all: My OS is 32bit winxp,but I wanna install 64bit R2.14.1. From the following website,it says You can also go back and add 64-bit components to a 32-bit install, or vice versa

Re: [R] question about rev

Yes£¬I find out later. rev is only reverse the order. What I use is ! Thanks At 2012-01-06 16:07:11,Petr PIKAL petr.pi...@precheza.cz wrote: Hi [R] question about rev Hi,all: I have a vector,and wanna get the opposite value via rev function. a [1] FALSE FALSE TRUE TRUE

[R] question about rev

Hi,all: I have a vector,and wanna get the opposite value via rev function. a [1] FALSE FALSE TRUE TRUE TRUE rev(a) [1] TRUE TRUE TRUE FALSE FALSE I don't know why the 3rd TRUE has not been reversed,while all other values are reversed? Thanks My best [[alternative

[R] My finding-resent

Sorry for some typo last mail. I corrected it,and resent.Sorry for it. Sir: I find out that for 2 level factor, if I set it to factor, then I'll get error reply. For the instance last mail, if I use: result1-glm(y ~ factor(gender),family = binomial);#gender has 2 levels

[R] How to compute 95%CI for OR from logistic regression?

Hi all: My data has 3 variables: age(3levels : 30y=1 30-50y=2, 50y=3) gender(Male=0, Female=1) CD4 cell count(raw lab measurement) y(1:death 0:alive) I perform logistic regression to find out the factors that influence y. result-glm(y ~ factor(age) + factor(gender) + CD4,family = binomial)

[R] Error

Hi sir: I follow your suggestion: result-glm(y ~ factor(age) + factor(gender) + CD4,family = binomial) logistic.display(result) Error in coeff[, 1] : incorrect number of dimensions At 2011-12-14 01:59:36,Jorge I Velez jorgeivanve...@gmail.com wrote: Hi there, Try

[R] New question

According to the example of logistic.display: model0 - glm(case ~ induced + spontaneous, family=binomial, data=infert) summary(model0) logistic.display(model0) induced: 3levels 0,1,2 spontaneous: 3levels 0,1,2 So if 0 is reference, we should get 2 OR for induced1, induced2, spontaneous1,

Re: [R] Error

Yes,I¡¡understand. Thanks for your help. At 2011-12-14 13:53:21,Virasakdi Chongsuvivatwong cvira...@gmail.com wrote: logistic.display need a rather tidy model ie all independent variables must be the original name not any function of a variable in the dataset of the model. If data1 is your

Re: [R] New question

Yes,it works well. Thanks for your help. At 2011-12-14 13:06:14,Jorge I Velez jorgeivanve...@gmail.com wrote: Hi lm_mengxin, If that's the case, just use as.factor(): fit - glm(case ~ as.factor(induced) + as.factor(spontaneous), family=binomial, data=infert) logistic.display(fit)

[R] My finding

Sir: I find out that for 2 level factor, if I set it to factor, then I'll get error reply. For the instance last mail, if I use: result1-glm(y ~ gender,family = binomial);#gender has 2 levels logistic.display(result1) Error in coeff[, 1] : incorrect number of dimensions if I use:

[R] The test of randomized slopes(intercepts)

Hi all: I perform the linear mixed model for 300 persons, y is CD4 count,x is time. I randomized slope and intercept,so I can get 300 slopes and 300 intercepts.Now I wanna test wheter the variance of 300 slopes and 300 intercepts differs from zero. If the variance of 300 slopes(or intercepts)

[R] question about the axis label

Hello sir: I have a question about the axis label of scatterplot3d function. The data is in the attachment. If I use the command: scatterplot3d(x,y,z,type=h) I want the plot's x-axis lab to be 1,2,3,...,13, y-axis lab to be 1,2,3,...11 But if I use the command:

[R] about the trend comparison by R

Hi all: I've 2 groups of data,which are measured at: 1990,1991,1992,...2000. I wanna know whether the trend of 2 groups are same or not. Maybe time series analysis is suitable,but I don't know which index is fit for my purpose,and how to get the index from function of time series analysis

[R] How to draw multi group plot?

As the attachement,I wanna draw multi group plot. But I can only use : plot(x,y...) points(...) It's a heavy work to use these command if there're too many groups to be drawn because I have to use point() for many times. I wanna know wheter there's command which can draw the multigroup

[R] about heatmap

Hi all: As to the heatmap function, the default style is red and yellow,and red refers to low level and yellow refers to high level. How can I change the style to the contrary: red refers to high level and yellow refers to low level? Thanks a lot! My best [[alternative HTML version

[R] the problem about sample size

Hi all: I am a user of JM package. Here's the problem of sample size. The warning is: Error in jointModel(fitLME, fitSURV_death, timeVar = time, method = piecewise-PH-GH) : sample sizes in the longitudinal and event processes differ. According to the suggestion of missing data,I use the

[R] a question about Joint Model(JM)

Hell sir: I'm a user of JM and have a quesion. As to JM: An R Package for the Joint Modelling of Longitudinal and Time-to-Event Data, on page14: fitSURV - coxph(Surv(Time, death) ~ drug, data = aids.id, x = TRUE) I don't understand why use the dataset aids.id, but notaids as follow: fitSURV -

[R] a question about the command followup.plot of epicale package

Hi all: I have a question about the command followup.plot of epicale package. As to the demo data Orthodont, the command followup.plot works well.But if I delete some rows of data(delete Male data,and keep Female data only, for instance),the command can't work,and the warning is In

[R] a question about the variance of growth model

Hi all: In the growth model (Ref.Multilevel Modeling in R (2.3)),I have a question about the variance: On p.73, it says:To model decreasing variance one can use the varExp option. In cases where variance increases can use the varFixed option (see Pinheiro Bates, 2000 for details). My

[R] what ~1stands for?

Hi all: In the expression: lme(y~1,random=~1|time) What ~1 stands for? Thanks a lot! [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide

[R] a question about multilevelmodel

Hello all: I've read the document named A Brief Introduction to R, the multilevel package and the nlme package. At p68, one can transform the dataset to the required format by using make.univ. I wanna know,how the new variable MULTDV is calculated(can you show me the formula if possible

[R] about the cox result

Hi all: I finished cox analysis like this: fit_cox-coxph(Surv(dat$Time, dat$death) ~ dat$CD4 + strata(dat$gender),data=dat); fit_cox Call: coxph(formula = Surv(data_ori$Time, data_ori$death) ~ data_ori$drug + strata(data_ori$gender), data = data_ori) coef exp(coef)