search for: stephenc

Hi I am using boosting for a classification and prediction problem. For some reason it is giving me an outcome that doesn't fall between 0 and 1 for the predictions. I have tried type="response" but it made no difference. Can anyone see what I am doing wrong? Screen output shown below: > boost.model <- gbm(as.factor(train$simNuance) ~ ., # formula +

Hi I am doing something like this: hist(maximumPitch, xlab="Maximum Pitch in Hertz") which produces a nice histogram but what do I do to get two or three, etc on one page? I want to save the resulting file to an eps. I can find: postscript("ex.eps") which I then run something like my hist above and then dev.off() but I don't get anything in my

Hi I hope this isn't off topics, but I have always found when I stepAIC() some glm I get an improvement in accuracy and kappa, but I have just done a case where I got a marginal deterioration. Is this possible, or should I be going through my figures carefully to see if I have messed up? Stephen Choularton 02 9999 2226 0413 545 182 -------------- next part --------------

Hi I am just starting to look at R and trading in futures, stock, etc Can anyone point me to useful background material? Stephen Choularton 02 9999 2226 0413 545 182 -- 11:39 PM [[alternative HTML version deleted]]

Is there some decision tree method available with R, like ripper, that ends up producing a list of the rules and can be used for prediction? Stephen Choularton 02 9999 2226 0413 545 182 -- 5:40 PM [[alternative HTML version deleted]]

Hi I am trying to use tree() to classify movements in a futures contract. My data is like this: diff dip dim adx 1 0 100.00000 8650.0000 100.00000 2 0 93.18540 2044.5455 93.18540 3 0 90.30995 1549.1169 90.30995 4 1 85.22030 927.0419 85.22030 5 1 85.36084

...argument for passing new data is actually 'newdata', as in: > pred = predict(glm.model, newdata=form[150001:200000,-1], > type="response") Cheers Joe -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of stephenc at ics.mq.edu.au Sent: Tuesday, 16 October 2007 5:30 AM To: r-help at stat.math.ethz.ch Subject: [R] partitioning data I am trying to train on part of my data and test on another part: > glm.model = glm(as.factor(h_finished) ~ . , family=binomial, data=form[1:150000,]) > pred = predict(glm...

Hi can anyone recommend a perl module that I can use to run R? Stephen

Hi Everyone Thanks to those who responded last time. I am still having problems. I really want to find one of those tutorials on how to use svm() so I can then get going using it myself. Issues are which kernel to choose, how to tune the parameters. If anyone know of a tutorial please let me know. Stephen [[alternative HTML version deleted]]

Hi Does anyone know what this means: > glm.model = glm(formula = as.factor(nextDay) ~ ., family=binomial, data=spi[1:1000,]) > pred <- predict(glm.model, spi[1001:1250,-9], type="response") Warning message: prediction from a rank-deficient fit may be misleading in: predict.lm(object, newdata, se.fit, scale = 1, type = ifelse(type == 9 is my determinant and I still get

Hi I am just starting to look at R and trading in futures, stock, etc Can anyone point me to useful background material?

I am having a couple of problems someone may be able to cast some light on. Question 1: I am making a logistic model but when i do this: glm.model = glm(as.factor(form$finished) ~ ., family=binomial, data=form[1:150000,]) I get this: Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : variable lengths differ (found for 'barrier') which is

Hi I am using boosting for a classification and prediction problem. For some reason it is giving me an outcome that doesn't fall between 0 and 1 for the predictions. I have tried type="response" but it made no difference. Can anyone see what I am doing wrong? Screen output shown below: > boost.model <- gbm(as.factor(train$simNuance) ~ ., # formula +

I am trying to train on part of my data and test on another part: > glm.model = glm(as.factor(h_finished) ~ . , family=binomial, data=form[1:150000,]) > pred = predict(glm.model, data=form[150001:200000,-1], type="response") > t = table(pred, form[150001:200000,1]) Error in table(pred, form[150001:2e+05, 1]) : all arguments must have the same length but try as I might

Hi Everyone I am struggling to get going with support vector machines in R - smv() and predict() etc. Does anyone know of a good tutorial covering R and these things? Stephen [[alternative HTML version deleted]]

Hi I am trying to tune an svm by doing the following: tune(svm, similarity ~., data = training, degree = 2^(1:2), gamma = 2^(-1:1), coef0 = 2^(-1:1), cost = 2^(2:4), type = "polynomial") but I am getting Error in svm.default(x, y, scale = scale, ...) : wrong type specification! > I have to admit I am not sure what I am doing wrong. Could anyone tell me why the

Hi I can do this: formula = as.factor(outcome) ~ . in glm and other model building functions. I think there is a way to get the product of the determinants (that is d1 * d2, d1 * d3, etc) and also another way to get all the polynomials (that is like poly(d1,2) would produce for a single determinant). Can anyone tell me how you write them? Stephen [[alternative HTML version deleted]]

-----Original Message----- From: Stephen Collier Sent: Thursday, 2 October 2003 10:26 AM To: 'Gerald (Jerry) Carter' Subject: RE: [Samba] Samba 3.0.0 rpms Jerry, Thanks for the prompt reply. I obtained them from us1.samba.org and au1.samba.org I tried from both sites. It seems so strange as rc2, rc3 and rc4 installed perfectly. We were getting problems with them with current (up2date

I am trying to use bagging like this: > bag.model <- bagging(as.factor(nextDay) ~ ., data = spi[1:1250,]) > pred = predict(bag.model, spi[1251:13500,-9]) There were 25 warnings (use warnings() to see them) > t = table(pred, spi[1251:13500,9]) > t pred 0 1 0 42 40 1 12 22 > classAgreement(t) but I get the warning. The warnings run like this: >

Hi I am doing this sort of thing: POLY: > > obj = best.tune(svm, similarity ~., data = training, kernel = "polynomial") > summary(obj) Call: best.tune(svm, similarity ~ ., data = training, kernel = "polynomial") Parameters: SVM-Type: eps-regression SVM-Kernel: polynomial cost: 1 degree: 3 gamma: 0.04545455 coef.0: 0