similar to: Orthogonal regression

DeaR R-useRs, I'm trying to fit a logist model with these data: > dati y x 1 1 37 2 1 35 3 1 33 4 1 40 5 1 45 6 1 41 7 1 42 8 0 20 9 0 21 10 0 25 11 0 27 12 0 29 13 0 18 I use glm(), having this output: > g<-glm(y~x,family=binomial,data=dati) Warning messages: 1: Algorithm did not converge in: glm.fit(x = X, y = Y, weights = weights, start = start, etastart =

Hi, I suggest to give a look to: ?Practical Regression and Anova using R? by Julian Faraway http://cran.r-project.org/doc/contrib/Faraway-PRA.pdf http://www.stat.lsa.umich.edu/~faraway/book/ see also package faraway for datasets: http://cbio.uct.ac.za/CRAN/src/contrib/Descriptions/faraway.html for some econometric data sets: http://www.oswego.edu/~kane/econometrics/data.htm for data sets:

Dear UseRs, maybe is a silly question: how can I get Empirical CDF values from an object created with ecdf()?? Using print I obtain: Empirical CDF Call: ecdf(t) x[1:57] = 4.1, 4.4, 4.5, ..., 491.3, 671.27 Thanks in advance. Regards, Vito Diventare costruttori di soluzioni Became solutions' constructors "The business of the statistician is to catalyze the scientific

Hi, you can address to a single ts in a multivariate ts object by namets[,index]. See this example: > dati X Y 1 100 200 2 150 210 3 180 220 4 200 230 5 220 250 > serie<-ts(dati,start=1999) > serie Time Series: Start = 1999 End = 2003 Frequency = 1 X Y 1999 100 200 2000 150 210 2001 180 220 2002 200 230 2003 220 250 > serie[,1] ## first ts Time Series: Start =

Dear R-useRs, I've written a contribute (in Italian language) concering fitting distribution with R. I believe it could be usefull for someones. It's available on CRAN web-site: http://cran.r-project.org/doc/contrib/Ricci-distribuzioni.pdf Here's the abstract: This paper deals with distribution fitting using R environment for statistical computing. It treats briefly some

Hi, I've a list containing parameters (intercepts & coefficients) of 12 regressions fitted > coeff [[1]] (Intercept) anno -427017.1740 217.0588 [[2]] (Intercept) anno -39625.82146 21.78025 ..... [[12]] (Intercept) anno 257605.0343 -129.7646 I want create a data frame with two columns (intercept and anno)using data in these list. Any help

Hi, I'm dealing with time series. I usually use stl() to estimate trend, stagionality and residuals. I test for normality of residuals using shapiro.test(), but I can't test for autocorrelation and heteroskedasticity. Is there a way to perform Durbin-Watson test and Breusch-Pagan test (or other simalar tests) for time series? I find dwtest() and bptest() in the package lmtest, but it

Hi, in which R-package I could find skewness and kurtosis measures for a distribution? I built some functions: gamma1<-function(x) { m=mean(x) n=length(x) s=sqrt(var(x)) m3=sum((x-m)^3)/n g1=m3/(s^3) return(g1) } skewness<-function(x) { m=mean(x) me=median(x) s=sqrt(var(x)) sk=(m-me)/s return(sk) } bowley<-function(x) { q<-as.vector(quantile(x,prob=c(.25,.50,.75)))

Hi, I found this document, but it concerns S+. If it could interest you'll see: http://faculty.washington.edu/ezivot/book/QuanCopula.pdf Cordially Vito You wrote: Dear R: Is there a function or a reference to simulate Gumbel copulas, please? Thanks in advance! Sincerely, Erin Hodgess mailto: hodgess at gator.uhd.edu R version 2.0.1 windows ===== Diventare costruttori di soluzioni

Dear R-Users, How can I use chisq.test() as a goodness of fit test? Reading man-page I?ve some doubts that kind of test is available with this statement. Am I wrong? X2=sum((O-E)^2)/E) O=empirical frequencies E=expected freq. calculated with the model (such as normal distribution) See: http://www.itl.nist.gov/div898/handbook/eda/section3/eda35f.htm for X2 used as a goodness of fit test. Any

Dear R-users, I'm finding for a R-package concerning graphs. Is there some kind of that package? I've a set of correlation coeffients between several variable and I wish to built a graph to link variables correlated. Many thanks. Best, Vito ===== Diventare costruttori di soluzioni "The business of the statistician is to catalyze the scientific learning process." George

I hope this example could help you best vito > x<-seq(-3.5,3.5,0.1) > x [1] -3.5 -3.4 -3.3 -3.2 -3.1 -3.0 -2.9 -2.8 -2.7 -2.6 -2.5 -2.4 -2.3 -2.2 -2.1 [16] -2.0 -1.9 -1.8 -1.7 -1.6 -1.5 -1.4 -1.3 -1.2 -1.1 -1.0 -0.9 -0.8 -0.7 -0.6 [31] -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 [46] 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3

Hi! I am searching for the Cochran-Armitage-trend-test. Is it included in an R-package? Thank you! --

When using fitdistr() with the exponential, log-normal and beta distributions, you get the relevent rate, mean, standard deviation, shape1 and shape2 but you get a number bellow those that are in () and I was wandering what exactly those numbers represent and how they relate to the data. Many thanks Mark Miller

Dear all, I'm looking for some material on data mining with R. I have something from Luis Torgo but I'd like to see something else. If anybody could help me I'll be thankful Adri??n

Hello, anyone got an idea on how to use stl() so that the remainder eventually becomes white noise? i used stl repeatedly but there is autocorrelation in the remainder that i can't get rid of. os: linux suse9.3 ------------------------------------------------ Sebastian Leuzinger Institute of Botany, University of Basel Sch??nbeinstr. 6 CH-4056 Basel ph 0041 (0) 61 2673511 fax 0041 (0)

hi all i have a another stats question. i would like to solve the following question: y(i)=a+b*x(i)+e(i) i.e. estimate a and b (they should be fixed) but i dont want to specify the standard density to the straight line. this can be done using kernel regression. the fitted line is however fitted locally. does anyone have a reference that will help me with my problem. i am still new to

Hi, my name is Giacomo. I would like to know how to create a Trend variable in a regression using R. Thank you for your help. My best regards, Giacomo --------------------------------- [[alternative HTML version deleted]]

Hello to everybody, I'd like to submit a problem I'm dealing with, and I can't get an answer to by myself. I have to test if my data come from a specific probability distribution, of which I know the analytic form both of the p.d.f. and the c.d.f. Namely, it is the hypoexponential distribution, sum of two exponentials with different parameters. Is there any way I can manage the

Does p-value on Box.test(data,lag=l) returns probability, that H0: cor(1)=cor(2)=..=cor(l)=0 holds? Thanks. [[alternative HTML version deleted]]