Search String: Display: Description: Sort:

Results:

References: [ +from:phlasky@earthlink.net: 27 ]

Total 27 documents matching your query.

1. Find the K largest & smallest entries in a Correlation Matrix (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Sat, 4 Jun 2005 08:20:19 -0700
How do I find the k largest and smallest entries in a large correlation matrix and their addresses ( excluding the main diagonal entries of 1?) For example, given the correlation matrix for all SP500
/archives/html/s-news/2005-06/msg00011.html (8,801 bytes)

2. Find the Largest and Smallest Entries in a Correlation Matrix (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Wed, 8 Jun 2005 07:07:25 -0700
Thanks to Patrick Burns of Burns Statistics who suggested use of the function lower.tri( ). This function extracts the lower Part of the matrix and renders it a vector. So to effectively find the col
/archives/html/s-news/2005-06/msg00028.html (12,323 bytes)

3. Calling rpart from a function (score: 1)
Author: o.au>
Date: Thu, 14 Apr 2005 17:13:10 -0700
6:54 AM To: 's-ne
/archives/html/s-news/2005-04/msg00106.html (8,120 bytes)

4. Eliminating NA's in a Time Series (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Fri, 13 Aug 2004 08:12:03 -0700
How do I eliminate NA's that are randomly scattered through-out a 2 column timeSeries ( dataSeries(TS) yields a 2 column matrix)? The new timeSeries should "drop" non-NA positions in each column that
/archives/html/s-news/2004-08/msg00080.html (6,922 bytes)

5. Eliminating NA's in a Time Series (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Tue, 17 Aug 2004 12:24:45 -0700
d one is up
/archives/html/s-news/2004-08/msg00092.html (8,783 bytes)

6. S+ version of quadprog ? (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Wed, 12 May 2004 08:27:39 -0700
Does anyone know of a Splus (version 6.x - Win Xp system) version of quadprog ? All I can find is a R version at CRAN and this doesn't work with Splus. Paul Lasky P & B Consultants Rancho Santa Fe, C
/archives/html/s-news/2004-05/msg00059.html (6,870 bytes)

7. Non-Linear Optimizers --- A comparison with a Stochastic Programming problem. (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Mon, 31 May 2004 20:21:08 -0700
I've recentally had an opportunity to compare and stress the 2 available non-linear optimizers available to S+ and R users: MASS's optim, and Splus's nlminb. In actual practice, optim far out perform
/archives/html/s-news/2004-05/msg00163.html (9,311 bytes)

8. 3 Dimensional Cross Variance (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Thu, 22 Apr 2004 16:25:34 -0700
This is a pretty basic question but . . . is there a better way, without explicit looping, to compute the 3 dimensional cross variance and a diagonal ? The crossvariance defined as: crossvar = array(
/archives/html/s-news/2004-04/msg00105.html (9,868 bytes)

9. Density Estimation in p dimensions (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Fri, 13 Feb 2004 09:39:19 -0800
This is not a strictly S+ or R problem but rather a statistical estimation problem - - - perhaps someone has some idea of how to solve the problem. I have a p dimension vector where p maybe as large
/archives/html/s-news/2004-02/msg00106.html (7,826 bytes)

10. Density Estimation in p dimensions (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Sat, 14 Feb 2004 12:29:41 -0800
Many thanks to all that replied. However no practical solutions were offered, so I devised one that seems to work very well on my data set. Several people stated that the problem was "very difficult"
/archives/html/s-news/2004-02/msg00113.html (11,290 bytes)

11. r (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Thu, 22 Jan 2004 09:15:45 -0800
How do I generalize permutation testing? Following up on the permutation testing question: I have a 3 factors (e.g. Size, Style, and Family) and each factor has 3 levels (e.g Size: large, mid, and sm
/archives/html/s-news/2004-01/msg00116.html (9,260 bytes)

12. MARS (Multivariate Adapative Regression Splines) (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Mon, 15 Dec 2003 07:50:45 -0800
Does anyone know of source for a Splus MARS (Multivariate Adapative Regression Splines) routine ? Paul Lasky phlasky@earthlink.ne "Truth is embodied in successful prediction"
/archives/html/s-news/2003-12/msg00079.html (6,827 bytes)

13. Color of a Trellis Box Plot (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Mon, 22 Dec 2003 15:16:06 -0800
teg
/archives/html/s-news/2003-12/msg00124.html (7,486 bytes)

14. Color of a Trellis Box Plot (score: 1)
Author: "Paul Lasky" <phlasky@earthlink.net>
Date: Wed, 24 Dec 2003 08:38:10 -0800
ng valu
/archives/html/s-news/2003-12/msg00129.html (9,364 bytes)

15. inverting a timeDate sequence (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Tue, 16 Sep 2003 12:36:28 -0700
How do you create a timeDate sequence with the latest date as the first element and oldest date the last element, AND exclude weekends (utilizing Splus 6.1)? The following efforts were unsuccessful:
/archives/html/s-news/2003-09/msg00108.html (8,586 bytes)

16. Re: inverting a timeDate sequence Summary of Replies (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Thu, 18 Sep 2003 10:19:42 -0700
How do you create a timeDate sequence with the latest date as the first element and oldest date the last element, AND exclude weekends (utilizing Splus 6.1)? The following efforts were unsuccessful:
/archives/html/s-news/2003-09/msg00120.html (9,882 bytes)

17. Re: importance sampling (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Sat, 16 Aug 2003 07:19:14 -0700
Importance sampling is simply weighted sampling with weights equal to dG/dH, where H is the sampling PDF and G is the desired or target PDF. You can easily accomplished weighted sampling with sample
/archives/html/s-news/2003-08/msg00084.html (7,868 bytes)

18. Replace apply with colMeans, etc (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Fri, 11 Jul 2003 09:19:35 -0700
In building a simulation program in S-Plus 6.1, windows, I must calculate some column means of a matrix (NOT a data frame) (many times) x 10 ^ 5. I was surprised to find that the function colMeans is
/archives/html/s-news/2003-07/msg00082.html (7,650 bytes)

19. Time Series: How to build irregular sequences without tears (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Mon, 23 Jun 2003 07:45:18 -0700
There is no "c" or concetration function for timeDate objects. And the append function cannot be used to splice together two timeDate sequences. How then do you build irregular time series, e.g. how
/archives/html/s-news/2003-06/msg00158.html (9,083 bytes)

20. Re: Recreating correlation (score: 1)
Author: "Paul H. Lasky" <phlasky@earthlink.net>
Date: Wed, 25 Jun 2003 09:36:13 -0700
This is a common task in simulating the end-point of financial derivatives. Assume that you know the distribution of each component of A, call it Phi ( assume that the distribution all components is
/archives/html/s-news/2003-06/msg00181.html (10,397 bytes)


This search system is powered by Namazu