Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Fri, 2 Mar 2001 09:20:09 +0100 (MET)
I have a question that is not originally related to S+, but I would like to ask to this specific community. Please do not reply to the list, but to my original email address. We are going to compare
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Tue, 24 Oct 2000 12:19:35 +0200 (MET DST)
Hello together, I know that there is no definitely 'right' way of the selection of predictors in regression models, but one small question: I fitted a logistic regression model and selected predictor
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Tue, 15 Aug 2000 09:46:10 +0200 (MET DST)
following model for 21 donors giving funds (y) in 4 different years, using another continuous variable (x) as a predictor: y(it) = b0 + b1x(it) + e(it); e(
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Thu, 17 Aug 2000 10:16:23 +0200 (MET DST)
with vertical bars, and error bars? (Splus 2000, Windows 95). Sorry if this has been asked before. Cheers, Blair -- This message was distributed by s-news@
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Tue, 4 Apr 2000 10:56:17 +0200 (MET DST)
Thanks for all who responded to my question!! It seems there are 3 basic strategies for dealing with heterogeneity in variance: a) transform the dependent variable (eg. log(y), or Box-Cox transformat
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Fri, 7 Apr 2000 15:21:22 +0200 (MET DST)
Hello everybody, I am trying to plot a trellis panel with nested data. I want each panel to have 3 differently specified lines in it (like the pixel-example [figure two] in the paper from Pinheiro an
Author: Matthias Richard <richard@psyres-stuttgart.de>
Date: Tue, 18 Jan 2000 09:30:25 +0100 (MET)
Hi Nicole, I had the same problem and 'solved' it by using the lines() function to put a line across the space that whas overwritten by the key() function. Serious drawback: it costs time to try out
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Thu, 25 Nov 1999 14:19:11 +0100 (MET)
Hello together, I have set up a gam()-model and would like to test a set of variables (each individually) in interaction with some factors in the model. To save a lot of typing and comparing I would
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Tue, 19 Oct 1999 14:01:31 +0100 (MET)
Hello folkes, I have a question concerning a 'simple' logistic regression model with two dummy coded effects (one with 4 categories in 3 dummies and the other with 2 categories in 1 dummy) and the in
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Thu, 21 Oct 1999 09:20:21 +0100 (MET)
Frank Harrell's response was the only one I got (see below). Thanks!!! yours, Matthias -- Matthias Richard (doctoral candidate) Center for Research on Psychotherapy Christian-Belser-Str. 79a 70593 Su
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Thu, 28 Oct 1999 08:22:37 +0100 (MET)
Hello all, I have a question concerning the trellis panel boxplots. I would like to draw series of _vertical_ boxplots in different panels and connect the medians within each panel with a line. Two q
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Fri, 24 Sep 1999 13:10:33 +0200 (MET DST)
This is a question concerning the bootstrap function of the HMISC-library written by Harrell (I use SPLUS 3.3 under windows95). I am interested in the corrected R-square of a final fitted model (logi
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Tue, 6 Jul 1999 18:36:26 +0200 (MET DST)
Hello everybody, I do a discrete time survival analysis and I would like to plot the (logit) hazards together with their confidence limits. I use the glm() as well as the gam()[with splines] function
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Thu, 15 Jul 1999 13:18:11 +0200 (MET DST)
Hello everybody! I would like to fit a gam()-model WITHOUT the intercept term. I tried to do it with -1 at the end of the formular, but it does not work. e.g. object1<-gam(EVENT ~ SP1INT + s(SP1MON)
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Fri, 16 Jul 1999 18:14:43 +0200 (MET DST)
Hello everybody! I am trying to understand the predict()-function for gam.objects. I am fitting two independent smoothing functions for two groups and want to create predictions using predict() with
Author: Matthias Richard <richard@Psyres-Stuttgart.DE>
Date: Tue, 23 Feb 1999 16:15:35 +0100 (MET)
Hello together, here is the summary of the responses to my question. function: Function multinom in library nnet. download and literature: http://www.stats.ox.ac.uk/pub/MASS2 is the nearest site to y