similar to: Fwd: Re: Prediction using GAM

Recently I was using GAM and couldn't help noticing the following incoherence in prediction: > data(gam.data) > data(gam.newdata) > gam.object <- gam(y ~ s(x,6) + z, data=gam.data) > predict(gam.object)[1] 1 0.8017407 > predict(gam.object,data.frame(x=gam.data$x[1],z=gam.data$z[1])) 1 0.1668452 I would expect that using two types of predict arguments

Note: forwarded message attached. --------------------------------- New Yahoo! Messenger with Voice. Call regular phones from your PC for low, low rates. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://dovecot.org/pipermail/dovecot/attachments/20060328/7853fd34/attachment-0001.html> -------------- next part -------------- An embedded message was

Note: forwarded message attached. In addition to what I've already posted, the output from gdb when I try to use mutt is: Program received signal SIGABRT, Aborted. 0x18234dcf in kill () from /lib/libc.so.5 And then it dies. __________________________________ Do you Yahoo!? All your favorites on one personal page ? Try My Yahoo! http://my.yahoo.com -------------- next part

Anybody ? Note: forwarded message attached. Applications From Scratch: http://appsfromscratch.berlios.de/ ____________________________________________________________________________________ Need a quick answer? Get one in minutes from people who know. Ask your question on www.Answers.yahoo.com -------------- next part -------------- An embedded message was scrubbed... From: Sergei

This is in regards to the suggested use of type="lpmatrix" in the documentation for mgcv::predict.gam. Could one not get the same result more simply by using type="terms" and interpolating each term directly? What is the advantage of the lpmatrix approach for prediction outside R? Thanks. -- View this message in context:

I fitted a GAM model with Poisson distribution using the function gam() in the mgcv package. My model is of the form: mod<-gam(y~s(x0)+s(x1)+s(x2),family=poisson). To extract estimates at a specified set of covariate values I used the gam `predict' method. But I want to get estimate and standard error of the difference of two fitted values. Can someone explain what should I do? Thank

I am fitting a two dimensional smoother in gam, say junk = gam(y~s(x1,x2)), to a response variable y that is always positive and pretty well behaved, both x1 and x2 are contained within [0,1]. I then create a new dataset for prediction with values of (x1,x2) within the range of the original data. predict(junk,newdata,type="response") My predicted values are a bit strange

Hello, Is it possible to estimate prediction interval using GAM? I looked through ?gam, ?predict.gam etc and the mgcv.pdf Simon Wood. I found it can calculate confidence interval but not clear if I can get it to calculate prediction interval. I read "Inference for GAMs is difficult and somewhat contentious." in Kuhnert and Venable An Introduction to R, and wondering why and if that

Hi Gam in mgcv package is predicting negative values which should not be the case despite all the predictors and response variables are positive. Tried to use log link function but it did not help. Please help sunil -- View this message in context: http://www.nabble.com/negative-prediction-by-gam-%28mgcv-package%29-tp20494965p20494965.html Sent from the R help mailing list archive at

As I understand it,  predict.lm(l ,newdata=nd ,interval="confidence") yields confidence bands for the predicted mean of new observations and lm.predict(l ,newdata=nd ,interval="prediction") yields confidence bands for new observations themselves, given an lm object l.   However with regard to {mgcv} although  predict.gam (g ,se.fit=TRUE ,interval= "prediction")

I'm looking for the best way to do the following: run a set of GAM models, and then make predictions with new data. My problem is the size of the gam model object, I would like to strip it down to the bare minimum of information needed to apply the model to new data. For example, if this were a linear model, I would just keep the betas. If this were an ordinary spline fit, I think I

The documentation for predict.gam in library mgcv gives an example of using an "lpmatrix" to do approximate prediction via interpolation. However, the code is specific to the example wrt the number of smooth terms, df's for each,etc. (which is entirely appropriate for an example) Has anyone generalized this to directly generate code from a gam object (eg SAS or C code)? I wanted to

The plots look reasonable to me. The plot of residuals against linear predictor always looks scary when many of the fitted values are very close to zero, so I tend to look at residuals against sqrt(fitted) in such cases. I don't think that the presence of the zero curve is a reason to reject the model --- it's easy to produce such plots by fitting a completely correct model to simulated

Gam in mgcv package is predicting negative values which should not be the case despite all the predictors and response variables are positive. Tried to use log link function but it did not help. Please help ------------------------------------------------------------------------ ---------------- Sunil Kumar Sharma Australia ******************************* IMPORTANT MESSAGE

Dear R-friends, How do you test the goodness of prediction of a model, when you predict on a set of data DIFFERENT from the training set? I explain myself: you train your model M (e.g. glm,gam,regression tree, brt) on a set of data A with a response variable Y. You then predict the value of that same response variable Y on a different set of data B (e.g. predict.glm, predict.gam and so on).

-------------- next part -------------- An embedded message was scrubbed... From: Shigeki Misawa <misawa at bnl.gov> Subject: Re: Problem with keygen on Solaris 8 system. Date: Mon, 30 Apr 2001 23:59:00 -0400 (EDT) Size: 3533 Url: http://lists.mindrot.org/pipermail/openssh-unix-dev/attachments/20010501/a092310c/attachment.mht -------------- next part -------------- An embedded message was

-------------- next part -------------- An embedded message was scrubbed... From: Omayle.Rondon at movilnet.com.ve Subject: sftp Date: Tue, 5 Aug 2003 18:01:47 -0400 Size: 3494 Url: http://lists.mindrot.org/pipermail/openssh-unix-dev/attachments/20030806/50dbdc11/attachment.mht -------------- next part -------------- An embedded message was scrubbed... From: Omayle.Rondon at movilnet.com.ve

Dear R-helpers, I have the following data: y happenat x 5185 (07/22/05 00:05:14) 14 5186 (07/22/05 00:15:14) 14 5187 (07/22/05 00:25:14) 14 5188 (07/22/05 00:35:14) 14 ...... I want to choose between 07/25/05 15:30:00 and 07/26/05 12:30:00. Anybody had experience in handling this kind of data? Is there a simple way to subset by the variable

I've got a small problem. I have some observational data (environmental samples: abiotic explanatory variable and biological response) to which I've fitted both a multiple linear regression model and also a gam (mgcv) using smooths for each term. The gam clearly fits far better than the lm model based on AIC (difference in AIC ~ 8), in addition the adjusted R squared for the gam is

I'm fitting a Poisson gam model, say model<-gam(a65tm~as.factor(day.week )+as.factor(week)+offset(log(pop65))+s(time,k=10,bs="cr",fx=FALSE,by=NA,m=1),sp=c( 0.001),data=dati1,family=poisson) Currently I've difficulties in obtaining right predictions by using gam.predict function with MGCV package in R version 2.2.1 (see below my syntax).