Search String: Display: Description: Sort:

Results:

References: [ +from:alan.hochberg@prosanos.com: 36 ]

Total 36 documents matching your query.

1. Director, Biostatistics & Health Outcomes, ProSanos Corporation, Harrisburg, PA (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Mon, 27 Mar 2006 14:49:38 -0500
m
/archives/html/s-news/2006-03/msg00080.html (8,001 bytes)

2. Re: finding cutoff point for a continuous variable when outcome is binary. (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Wed, 25 Jan 2006 15:06:19 -0500
Run an ROC (receiver operating characteristic) analysis and see if your "continuous predictor" is any better than tossing a coin. See if the area under your ROC curve is significantly greater than 0
/archives/html/s-news/2006-01/msg00087.html (8,680 bytes)

3. Re: object.size (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 26 Jul 2005 14:18:05 -0400
I think that the function you gave ends up looking at object sizes for character vectors representing the names of your objects, or some such funky S-like thing. This seems to work:<o:p></o:p> <o:p>
/archives/html/s-news/2005-07/msg00140.html (11,076 bytes)

4. Re: Soliciting ideas [Fitting a formula to data from machines] (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 3 May 2005 10:22:11 -0400
The first thing to realize is that there are two sources of deviation between the machine results and the formula. The first is systematic error. In other words, the shape of the formula&#8217;s curv
/archives/html/s-news/2005-05/msg00010.html (13,758 bytes)

5. Re: Off-topic: Anocova of change scores with baseline (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 10 May 2005 13:34:07 -0400
+
/archives/html/s-news/2005-05/msg00039.html (14,424 bytes)

6. Re: Ripley revisited (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 12 May 2005 10:49:43 -0400
t
/archives/html/s-news/2005-05/msg00068.html (9,434 bytes)

7. Re: recreating a time series from a subset of fourier coefficients (score: 1)
Author: v.au>
Date: Fri, 29 Apr 2005 10:09:49 -0400
=T)/5) [1] 1.8673
/archives/html/s-news/2005-04/msg00177.html (8,496 bytes)

8. Re: How to change a data structure? (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Mon, 7 Feb 2005 08:46:26 -0500
I use this kind of question as a good exercise to train my mind out of for-loop thinking into the vectorized S-plus world. Here I was half-successful, vectorizing the columns but looping over rows. E
/archives/html/s-news/2005-02/msg00022.html (8,044 bytes)

9. Re: How to change a data structure? [Usefulness for clinical data] (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 8 Feb 2005 09:01:14 -0500
As we wrap this up, it's worth pointing out the importance of this particular problem (and its minor variants) in clinical data analysis. The Clinical Data Standards Interchange Committee (CDISC, www
/archives/html/s-news/2005-02/msg00032.html (8,461 bytes)

10. Re: Controlling colour coding in image plots (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Mon, 24 Jan 2005 14:45:14 -0500
f
/archives/html/s-news/2005-01/msg00121.html (8,464 bytes)

11. Re: hypergeometric distribution (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 27 Jan 2005 16:54:16 -0500
t
/archives/html/s-news/2005-01/msg00147.html (7,258 bytes)

12. Re: Cluster Analysis of Gridded Continuous Fields (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 23 May 2006 09:28:39 -0400
One of the difficulties of diagnosing a problem with clustering is that you have three sub-problems: 1) Choosing the right feature set 2) Choosing the right similarity measure 3) Choosing the right
/archives/html/s-news/2006-05/msg00084.html (11,216 bytes)

13. R or S-plus code for scaled rectangle diagrams? (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Fri, 2 Jun 2006 09:15:39 -0400
Does anyone know of an R or S-Plus function to draw scaled rectangle diagrams? This variation on Venn diagrams is described here:<o:p></o:p> <o:p> </o:p> Marshall, RJ. Displaying clinical data relati
/archives/html/s-news/2006-06/msg00001.html (10,788 bytes)

14. Re: automate writing formulas (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Mon, 31 Jul 2006 09:27:24 -0400
You could, if you wish, use names() to get the variable names in your data frame, use cat() with sink() to write them in the form of a formula into a little text file, and then copy/paste them into R
/archives/html/s-news/2006-07/msg00081.html (10,160 bytes)

15. Large-scale least-squares regression (was: automate writing formulas) (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Mon, 31 Jul 2006 12:41:06 -0400
Your point is well taken. When Acton's quote was written, your 300 variables might come from 300 sensors on an aircraft or a chemical reactor, and there was a chance that physical and engineering ins
/archives/html/s-news/2006-07/msg00083.html (8,218 bytes)

16. Re: automate writing formulas: No insult intended.... (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 3 Aug 2006 10:41:54 -0400
Dear Kamil,<o:p></o:p> <o:p> </o:p> Let me apologize publicly if you feel personally insulted. If I was warning you away from doing something I thought might be ill-advised, it&#8217;s only because I
/archives/html/s-news/2006-08/msg00009.html (9,124 bytes)

17. Re: Solving NonLinear Equations (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 7 Sep 2006 15:26:59 -0400
Note that your equations are non-linear but not inherently multivariate. They can be easily transformed to a univariate form that can be solved with the uniroot() function. The following is quick-an
/archives/html/s-news/2006-09/msg00017.html (8,762 bytes)

18. Re: Solving NonLinear Equations (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Thu, 7 Sep 2006 17:29:27 -0400
Uniroot() finds "a" solution in an interval, not necessarily "the" solution that you want. There may be multiple roots, some of which meet your constraints. Since you've got R or S-plus which have gr
/archives/html/s-news/2006-09/msg00020.html (8,079 bytes)

19. Re: 'Prompt to save graph sheets' as argument for options() (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Tue, 9 Jan 2007 11:51:36 -0500
<o:p> </o:p> Just about every command calls a function, and you can see these by using the &#8220;History&#8221; window. It looks like the command you need is:<o:p></o:p> <o:p> </o:p> guiSetOption(
/archives/html/s-news/2007-01/msg00009.html (11,831 bytes)

20. First Annual FDA/DIA Statistics Forum (score: 1)
Author: "Alan Hochberg" <alan.hochberg@prosanos.com>
Date: Wed, 7 Feb 2007 22:35:29 -0500
The 1st Annual FDA/DIA Statistics Forum is fast approaching. Program information and registration details can be found on the DIA website at:<o:p></o:p> <o:p> </o:p> http://www.diahome.org/DIAHome/Ed
/archives/html/s-news/2007-02/msg00018.html (13,665 bytes)


This search system is powered by Namazu