March 17, 2006

Coruscation Prospectus

Sitemap

Posted by omor at 10:12 PM | Comments (0)

October 15, 2005

Joint regression analysis

Joint regression analysis to study genotype-environmental interaction,
genotype effects and/or interaction effects within individual
environments are related to environmental effects.

The interaction sum of squares is divided into two parts:
* one part represents the heterogeneity of linear regression
coefficients while
* the second represents the pooled deviations from individual
regression lines.

R. J. (Bob) Baker

Posted by omor at 10:31 AM | Comments (0)

September 08, 2005

Hospital Length of Stay: Mean or Median Regression

Length of stay (LOS) is an important measure of hospital activity and
health care utilization, but its empirical distribution is often
positively skewed.

Median regression appears to be a suitable alternative to analyze
the clustered and positively skewed LOS, without transforming and
trimming the data arbitrarily.

Objective. This study reviews the mean and median regression
approaches for analyzing LOS, which have implications for service
planning, resource allocation, and bed utilization.

Methods. The two approaches are applied to analyze hospital discharge
data on cesarean delivery. Both models adjust for patient and
health-related characteristics, and for the dependency of LOS outcomes
nested within hospitals. The estimation methods are also compared in a
simulation study.

Results. For the empirical application, the mean regression results
are somewhat sensitive to the magnitude of trimming chosen. The
identified factors from median regression, namely number of diagnoses,
number of procedures, and payment classification, are robust to
high-LOS outliers. The simulation experiment shows that median
regression can outperform mean regression even when the response
variable is moderately positively skewed.

Conclusion. Median regression appears to be a suitable alternative to
analyze the clustered and positively skewed LOS, without transforming
and trimming the data arbitrarily.

Analyzing Hospital Length of Stay: Mean or Median Regression ?
Medical Care. 41(5):681-686, May 2003.
Lee, Andy H.; Fung, Wing K.; Fu, Bo
[**]

Posted by omor at 03:07 PM | Comments (0)

July 19, 2005

sas proc quantreg for quantile regression

Some PROC QUANTREG features are:

* Implements the simplex, interior point, and smoothing algorithms for
estimation

* Provides three methods to compute confidence intervals for the
regression quantile parameter: sparsity, rank, and resampling.

* Provides two methods to compute the covariance and correlation
matrices of the estimated parameters: an asymptotic method and a
bootstrap method

* Provides two tests for the regression parameter estimates: the Wald
test and a likelihood ratio test

* Uses robust multivariate location and scale estimates for leverage
point detection

* Multithreaded for parallel computing when multiple processors are
available

[PDF, *]

Posted by omor at 04:22 PM | Comments (0)

July 17, 2005

SAS examples with explanation at ucla.edu/stat/SAS/

SAS examples with explanation abound at UCLA: 1, 2.

Posted by omor at 10:38 PM | Comments (0)

July 14, 2005

sas proc sql joins data merging

data step merge statements corresponding to various joins.

INNER JOIN

merge data1 (in=a) data2 (in=b); by id;
if a and b;

LEFT JOIN

merge data1 (in=a) data2 (in=b); by id;
if a;

RIGHT JOIN

merge data1 (in=a) data2 (in=b); by id;
if b;

FULL JOIN
merge data1 (in=a) data2 (in=b); by id;

[ Quicktips 0206, A Visual Introduction to SQL Joins:, sql workshop at PennPop ]

Posted by omor at 04:56 PM | Comments (0)

July 12, 2005

sas proc sql assign value to macro variable

Calculate a value and assign it to a macro variable with
sas proc sql


proc sql noprint;
select ssn format=9. into :ssnok
separated by ' ' from ssn_list;
quit;

%put &ssnok;
123456789 234567890 345678901 456789012

[CREATING MACRO VARIABLES VIA PROC SQL PDF]

Posted by omor at 04:18 PM | Comments (0)

June 22, 2005

Good software documentation requires code formatting

Higher Order JavaScript has a decent sample of code formating
for documentation.

CSS:

pre {

	background-color: #e4f0e4;

	margin-left: 1em;

	border-top:  1px #d0d0d0 solid;

	border-left: 1px #d0d0d0 solid;

	padding: .6em 0;

}

What's the best way to represent SAS code ?

Posted by omor at 11:04 PM | Comments (0)

June 04, 2005

Bruce Gilsen, PROGRAM EFFICIENCY

Bruce Gilsen, Federal Reserve Board, offers advice on how
to program SAS efficiently.

[PDF]

Posted by omor at 08:06 PM | Comments (0)

June 01, 2005

two models in one regression

In SAS, you can estimate two distinct models with one call to proc reg.

proc reg data=USPopulation outest=est tableout alpha=0.1;
m1: model Population=Year/noprint;
m2: model Population=Year YearSq/noprint;
proc print data=est;
run;

*]

Posted by omor at 10:01 PM | Comments (0)

April 02, 2005

date simulation

Want a list of week-ending dates simulated, starting
at 2005 March 28 and ending at 2005 Nov 21.

Find magic number 16515 by trial and error,
or a SAS date function.

Find magic nbumber 250 by trial and error,
or a SAS date function.

data plan;
datebase = 16515;
do i = 1 to 250 by 7;
datex = datebase + i;
week = round (1+i/7);
output;
end;
format datex mmddyy10.;
run;
proc print data = plan; var datex week; run;

More SAS date dox.

Obs datex week

1 03/21/2005 1
2 03/28/2005 2
3 04/04/2005 3
4 04/11/2005 4
5 04/18/2005 5
6 04/25/2005 6
7 05/02/2005 7
8 05/09/2005 8
9 05/16/2005 9
10 05/23/2005 10
11 05/30/2005 11
12 06/06/2005 12
13 06/13/2005 13
14 06/20/2005 14
15 06/27/2005 15
16 07/04/2005 16
17 07/11/2005 17
18 07/18/2005 18
19 07/25/2005 19
20 08/01/2005 20
21 08/08/2005 21
22 08/15/2005 22
23 08/22/2005 23
24 08/29/2005 24
25 09/05/2005 25
26 09/12/2005 26
27 09/19/2005 27
28 09/26/2005 28
29 10/03/2005 29
30 10/10/2005 30
31 10/17/2005 31
32 10/24/2005 32
33 10/31/2005 33
34 11/07/2005 34
35 11/14/2005 35
36 11/21/2005 36

Posted by omor at 10:39 PM | Comments (0)

January 15, 2005

BLAST and SAS: String matching algorithms and their application

Was the author really Shakespeare?- String matching algorithms and
their application
Raymond Wan
Gilead Sciences, Inc.

Sting matching especially approximate (fuzzy) string matching is
important to a lot of different fields in computer science. It is
used in CRM, database cleaning, bibliometrics, and especially
bioinformatics. In fact, a large portion of the supercomputing
resources in the world is now devoted to an algorithm called BLAST
(Basic Local Alignment Search Tool), which is a fuzzy string matching
algorithm. At the heart of all these applications is the need to
measure how different two text strings are to each other. We will look
at two different ways to build a measure and how it can be
implemented in SAS.

June 2004 Bay Area SAS Users Group Meeting

When: 8 June 2004
Time: 1:00-5:00 p.m.
Where: Gilead Sciences, Inc.
320 Lakeside Drive, Foster City, CA 94404

Map
Host: Steve Wong
steve.wong@gilead.com

Contact/Information:
Glenn Itano (650-522-5671)
Sandy Chang (650-522-5285)

Posted by omor at 03:21 PM | Comments (0)

December 28, 2004

Rick Aster / SAS programming info

aka programming secrets:
Professional SAS Programming Shortcuts and Professional SAS Programming Logic.

Posted by omor at 09:51 PM | Comments (0)

December 25, 2004

SAS Proc Tabulate FAQ

SAS Proc Tabulate FAQ [ucla]
with a few axamples.

Posted by omor at 09:50 PM | Comments (0)

December 24, 2004

SAS ODS introduction

Use SAS to generate nice looking statistical report documents.

Excel (XLS) file

ods html file = "c:\temp\data.xls";
proc print data =new;run;
ods html close;

Web(HTML) file

ods html file = "body.html";
proc print data =new;run;
ods html close;

More at SAS ODS intro [PDF]

Posted by omor at 11:15 AM | Comments (0)

December 23, 2004

comp.soft-sys.sas SAS newsgroup

comp.soft-sys.sas SAS newsgroup at googlegroups.
Hands on statistical computing howto.

Posted by omor at 09:26 PM | Comments (0)

December 22, 2004

SAS blog

Weblogsinc's SAS blog is more about business than statistics.

Update 2005 Jan 15: Welcome Weblogsinc SAS blog readers.
There are more Coruscation SAS blog items.

Posted by omor at 09:12 PM | Comments (0)

December 11, 2004

SAS percentiles not automatically calculated ?

SAS tip: How do I obtain percentiles not automatically calculated ?

proc univariate data=hsb noprint;
var write;
output out=percentiles1 pctlpts=33 45 80 to 90 by 2 pctlpre=P;
run;
proc print data=percentiles1;run;

Posted by omor at 11:43 PM | Comments (0)

December 04, 2004

Histograms , superimposed with fitted probability density curves

PROC CAPABILITY is a component of SAS/QC (Quality Control). The
features described below are now available in PROC UNIVARIATE (part of base SAS).

# Histograms and comparative histograms. Optionally, these can be
superimposed with fitted probability density curves for various
distributions and kernel density estimates.

# Cumulative distribution function plots (cdf plots). Optionally,
these can be superimposed with specification limits and probability
distribution curves for various distributions.

# Quantile-quantile plots (Q-Q plots), probability plots, and
probability-probability plots (P-P plots). These plots facilitate the
comparison of a data distribution with various theoretical
distributions.

# Goodness-of-fit tests for a variety of distributions including the
normal.

# Statistical intervals (prediction, tolerance, and confidence
intervals) for a normal population.

# The ability to inset summary statistics and capability indices in
plots produced on a graphics device.

Posted by omor at 06:56 PM | Comments (0)

December 03, 2004

Paul Dickman

Paul Dickman has some good SAS tips for statistical programming
and handling datasets and simple graphics.

Posted by omor at 07:26 PM | Comments (0)

November 30, 2004

sconsig's SAS tips

SAS Consultants' Special Interest Group (sconsig)'s SAS tips,
a sample of SAS-L, SUGI, and support.sas.com.

Posted by omor at 02:09 PM | Comments (0)

November 29, 2004

SAS doc

SAS docs 9.1.2 (Official version)

Mirrors:
OK State SAS documentation.
Queens' SAS documentation.
Topics in SAS Programming [UNC]

Posted by omor at 03:52 PM | Comments (0)