Re: R vs. SAS (was Replacement for SAS (SPSS vs. SAS, redux)) #4
No, sorry if I made it sound that way.....
SAS and R are totally different in the way they work, how they are
structured, what the statements look like, and so on. Not like French
and Spanish, not even like English and Spanish. More like English and
R has a lot in common with S Plus, both are based on the S language.
But there is no hostility that I know of between S Plus and R (unlike
the situation with Linux)
>>> <ben.powell@CLA.CO.UK> 2/5/2004 10:36:32 AM >>> wrote
Sounds like R is the SAS equivalent of Linux .. who are incidentally in
sort...Re: R vs. SAS (was Replacement for SAS (SPSS vs. SAS, redux)) #2
Sounds like R is the SAS equivalent of Linux .. who are incidentally in all
sorts of a copy left/right mess at the moment.
On Thu, 5 Feb 2004 09:39:37 -0500, Peter Flom <flom@NDRI.ORG> wrote:
>I use both SAS and R; I"ve used SAS for 10 or 15 years, R for about 6
>months - 1 year.
>Each has strengths and weaknesses; I'd not like to give up either.
>Strengths of SAS (esp. compared to R)
>Tech support - I have SAS tech support's phone number and our license
>number memorized. They're great.
>SAS-L - There's also an R-help list...Re: R vs. SAS (was Replacement for SAS (SPSS vs. SAS, redux)) #3
Wow, that was a leap. Better not use Apache or PHP then.
>From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On
>Behalf Of ben.powell@CLA.CO.UK
>Sent: Thursday, February 05, 2004 10:37 AM
>Subject: Re: R vs. SAS (was Replacement for SAS (SPSS vs. SAS, redux))
>Sounds like R is the SAS equivalent of Linux .. who are
>incidentally in all
>sorts of a copy left/right mess at the moment.
>On Thu, 5 Feb 2004 09:39:37 -0500, Peter Flom <flom@NDRI.ORG> wrote:
>>I use both SAS a...Re: SAS vs. SPLUS vs. SAS #3
I don't know much about trials work, but it has been discussed on the r-help list, and there
are definitely people who use R for this type of thing
>From: Paul Miller <pjmiller_57@YAHOO.COM>
>Sent: Dec 11, 2009 4:09 PM
>Subject: SAS vs. SPLUS vs. SAS
>I’ve recently become interested in sequential clinical trials designs. I’ve purchased a book that discusses the topic called “Analysis of Clinical Trials using SAS: A Practical Guide.” Upon trying to run some of the ...Re: SAS vs. SPLUS vs. SAS 220046
If you are referring to matrix manipulations which can be
performed employing routines made available with PROC FCMP,
those routines are not an adequate substitute for a
full-fledged matrix programming language. Also, I have
found that the matrix routines are not really accessible
in SAS/STAT procedures the way that they should be.
I have brought this to the attention of SI. Their response
was that they were not aware of the problems until I reported
that I could not employ the matrix operations during execution
of NLMIXED code. On further consideration, they determined
that the...Re: SAS vs. SPLUS vs. SAS #4
matrix is indispensible tool in implementing modern data mining algorithm,
but we only have native matrix opertion capability in IML.
Recently I discovered that in SAS/Base SAS/Stat, we were able to conduct
several important matrix manipulations and I able to translate those
prototype algorithms on textbook into SAS.
I just submited a paper to SGF2010 on this topic, hope the paper can be
On Fri, 11 Dec 2009 13:09:32 -0800, Paul Miller <pjmiller_57@YAHOO.COM>
>I’ve recently become interested in sequential clinical trials des...Re: SAS vs. SPLUS vs. SAS #2 1617385
have not used S-Plus since around 2000-2002
so this may be Old News
from S-Plus conference attendance they know their strength is:
* rapid prototyping
* great graphics
* you want speed!
load all data into memory!
* need to upgrade hardware w/more memory for large data sets
* check documentation to find BigData functions
which, iirc, operate like SAS does: row-wise
previous version: v7
current version: v8.1
SAS/IML now has IML/Studio which enables access to R programs
R is the open-source version of S+
...SAS vs. SPLUS vs. SAS 149448
Hi Dale, Peter, Ron, Ken, Ajay, and "Oloolo",
Thanks for your replies to my questions about using SAS and SPlus for seque=
ntial trial design. As I said earlier, I feel that, as a data manipulation =
procedure,=A0IML ought to be included in Base SAS. So I don't think we're g=
oing to pay the extra $6400=A0for it ($6400=A0for the first year plus $1200=
every year thereafter). And=A0Seqdesign and Seqtest=A0probably won't be re=
ady for use in the "production environment" at any point in the near future=
..=A0So it looks like I need to find a non-SAS=A...Re: SPSS vs. SAS, redux #5
I agree that you need to be a "statistician" and a "programmer (of sorts)",
however let me express my 2 cents.
I am both a statistician and a programmer who uses both SPSS and SAS. I
find that the point and click method in SPSS sometimes helps save time. For
example, while I do code in both SPSS and SAS when using SPSS if I am doing
a simple analysis, I can run the point and click method and have SPSS
generate the code. I can then take that code, review it to ensure it is
doing what I want, copy it and past to a "syntax" file and then alter the
code in any way...Re: What r the data types in SAS? in Base SAS , and SAS SQL
> From: Amar Mundankar
> Sent: Tuesday, July 21, 2009 8:10 AM
> To: email@example.com
> Subject: What r the data types in SAS? in Base SAS , and SAS SQL
> Hi all,
> What are the different data types in Base SAS and SAS SQL??
character, lengths from 1 to 32,000+
numeric: lengths from 2 to 8
dates are a subtype of numeric
and are identified by their date, datetime, or time formats
the new proc TSPL (Table Server Processing Language)
supports ANSI data types: bigint, tinyint, etc.
http://s...Re: PC SAS vs. Mainframe SAS
If you go this route make sure you have SAS/Connect module as you will want
to more than likely rsubmit your code to the mainframe have it do the heavy
lifting and then just shoot theresults back to your PC. From my experience
I would have rather just stayed with the mainframe. YOu will need to teach
people the difference between server side and client side and how to write
and mod the existing programs accrodingly. Also you will need to be careful
with using macros and macro variables as you really have to understand when
things get resolved and when things get passed and how to pass a macro
variable from the client side to the server side.
As for a drain on your PC, not much unless you start donloading abunch of
really big datasets. Then yoru gonna eat your PC harddrive space.
When everything is coming at you all at once, your in the wrong lane.
A truly happy person is someone who can smile and enjoy the scenery on a
From: elsolo <elsolo21@YAHOO.COM>
Reply-To: elsolo <elsolo21@YAHOO.COM>
Subject: PC SAS vs. Mainframe SAS
Date: Fri, 15 Sep 2006 08:16:57 -0700
We currently use SAS on a mainframe running z/os. We are contemplating
moving to the SAS for Windows version. I'm trying to put together a
list of pros and cons.
For pros, I have simple things such as: Friendly user interface,
better reporting directly from sas, easier to export files, easier to
For Cons I have obv...WPS and R vs. SAS and SAS/STAT
I am looking at the price for getting SAS software. I will need SAS/
STAT and SAS/GRAPH as well. I can not afford (and do not want to pay)
the =806000 for the first year and the =802000 for each year after that so
I am thinking of getting WPS and R and the WPS to R bridge because it
is much cheaper. Has anyone tried to do this in earnest and what were
their findings? Is this a feasible combination for doing clinical
...SAS-L Panel / SAS-L vs. SAS Institute Tech Support
[This is part of a report on a panel discussion about SAS-L which took
place during the 2003 NESUG conference. Partcipants: Paul Dorfman, Ron
Fehd, Phil Mason, Barbara Okerson, Dianne Louise Rhodes, David Septoff;
also Ray Pass]
On the subject of SAS-L vs. SAS Institute Tech Support ...
Dianne noted that a real strength of SAS-L is that it functions 24
hours a day, seven days a week. She recalled a situation in which she
helped somebody who had a project deadline on Christmas Eve.
Ron pointed out that SI technical support sometimes is slower because
problems often must be handed off, somet...Re: SAS vs BUGS, R, SPSS, etc. #5
There is one paper from Guo et al (2000) where they have discussed
multilevel modeling with binary data . If you go to multilevel modeling
website http://www.cmm.bristol.ac.uk/ you will find a paper by Kelvin Jones
where they have discussed various softwares of multilevel modeling . Once
you read this you will have an idea which softwares you need for your
study. BUGS is concerned with Bayesian approach, you can download it free
but its totally bayesian multilevel modeling. You can also visit Judith
Singer webpage where you will find a ppt entirely devoted to multilevel
modeling softwares. I have tried almost all softwares except Mlwin & I must
admit SAS Glimmix is the best.
All the best !
On Sun, Jun 1, 2008 at 1:13 PM, Phil Rack <PhilRack@minequest.com> wrote:
> Well, I think you have the tools to find out! After all, R is free to
> download and use so cost is not a barrier. I personally don't enjoy using
> but I find a lot of people are using it and are able to accomplish some
> work. My personal opinion and preferences is to use SAS (or WPS) to prep
> data and do basic reporting. If I couldn't afford the SAS statistical
> modules, I would use R, BUGS (your posting is the first time I've ever
> of BUGS) or take a look at RAPID-I.
> Other than scoring the data set, do you really need to perform statistical
> calculations on all the observations? Can you sample from your data to
>...Re: PC SAS vs. Mainframe SAS #15
I disagree. There is no way that a mainframe is stronger than a grid. A
mainframe can be part of a grid and hence is a subset. As an example, a
mainframe cannot do search queries faster than Google.
The old comparison of a PC and a mainframe cost-wise is out of date. Today's
PCs can be extremely safe and maintainable, they are fast in every measure,
and they provide a much better user experience. Wall clock time, my guess is
that the jobs will also run faster. With quad-core PCs available by this
Christmas running virtualization, you could have an 8 way on the desk,
running RAID...Re: PC SAS vs. Mainframe SAS #24
I agree that the days of large mainframe purchases are nearing their
end, but for a slightly different reason - licensing costs.
MVS is very expensive compared to Linux, and doesn't make up for the
increased licensing fees by reducing the need for support personnel.
In the case of SAS, you can buy a server license that's just big enough
to suit your needs, instead of a license for a machine where 99% of the
users will never invoke SAS, directly or indirectly.
MVS is much better at resource control than Unix (and I think it's Unix
that's replacing MVS, not Windows), but it...Re: PC SAS vs. Mainframe SAS #30
Joe is right. In short, many years ago (in a gigantic mainframe shop) I
was asked to "look at" a tabulate-like report produced by 100,000+ lines
of Cobol code b/c something was wrong with the numbers. I am not a
Tabulate expert, so it took me a whole hour to write the Table statement
producing about the same as Cobol - only with correct figures (since it
was Tabulate doing the aggregation). SAS programs had been then
prohibited from production (Cobol-only), so using my output (together
with Tabulate formatting) as a template, two highly qualified Cobol
programmers were given a month ...Re: PHREG SAS 8 vs SAS 9
Sorry I changed the number of loops to 10 instead of 1000 to reduce the
running time. But still, it takes much longer in SAS 9 compared to SAS
----- Original Message -----
From: "Francois Torche" <firstname.lastname@example.org>
Sent: Fri, December 19, 2008 13:58
Subject:PHREG SAS 8 vs SAS 9
The following code runs in less than a minute on SAS 8.2 and more than 11
on SAS 9. I know it's something related to PHREG but I don't know why?
Any help would be appreciated ;-)
VALUE dis_fmt 1='Case' 0='Control';
VALUE exp_fmt 1='Vaccntd' 0='Unvaccntd';
DATA test (label = Output Case-Control Macro 19DEC08:13:18:12 Search B);
Ncases = 200;
C_C_Rat = 2;
Vac_Cov = 0.8;
Market_sh = 1;
OR = 0.5;
Alpha = 0.05;
_theta = 2.5;
Limit = 1;
Uplow = "U";
SET test nobs=last;
IF _N_ = last THEN CALL SYMPUT("NofTest",put(_N_,best.));
if (Vac_cov * OR)/(1-Vac_cov+Vac_cov * OR) > 1 then delete;
RD = (OR-1) * Vac_cov * (1-Vac_cov) / (1-Vac_cov + OR * Vac_cov);
Zalpha = PROBIT(1-Alpha/2);
LABEL Ncases = "Number of cases" C_C_rat = "Case-control ratio" VAC_COV =
"Vaccination coverage - baseline - P(vaccinated|not ill)" OR = "Odds
Ratio" Alpha = "Alpha
Error - Type I" Limit = "Upper or Lower limit for the confidence interval
of OR or VE" Uplow =
"Say if Limit is an upper or lower lim...Re: PC SAS vs. Mainframe SAS #16
Von: SAS(r) Discussion Im Auftrag von Alan Churchill
Gesendet: Montag, 18. September 2006 13:47
Betreff: Re: PC SAS vs. Mainframe SAS
Whale vs school of minnows is the popular analogy.
While our popular analogy is: How many Cessna's does it take to
replace a Boeing 747?
...Re: PC SAS vs. Mainframe SAS #34
In the late 80's, I watched an IT department refuse to give our
outlying lab enough computing power. So our lab put in some
unix boxes, and one of our guys installed a Wollongong setup so
that the unix network could talk with the VAX that was all we
were supposed to have. The IT guys flew out to argue with us
that unix was a bad decision.
They said, "You can't talk to the VAX systems!"
We said, "We do already. Look! It's already up and running!"
Their response: "No, you can't get that to work."
None are so blind as those who will not see.
...Re: PC SAS vs. Mainframe SAS #28
What was it recently? No camera phones. That didn't last long because they
are all now camera phones. I went onsite at one place that disabled USB
ports to block USB keys. Another place said that Visual Studios was a
security risk. On that one, they finally backed down when I simply said
It's non-ending with them. That's why a lot of companies outsource the
function because they hinder business.
Savian "Bridging SAS and Microsoft Technologies"
From: SAS(r) Discussion [mailto:SAS-L@LI...Re: PC SAS vs. Mainframe SAS #29
the Dilbert character "Mordac, Preventer of Information Technology"
I remember hearing so many stories of the bizzare "rules" some shops would adopt. Paul has a story about use of Tabulate at one shop he worked.
---- Original message ----
>Date: Tue, 19 Sep 2006 11:13:02 -0600
>From: Alan Churchill <SASL001@SAVIAN.NET>
>Subject: Re: PC SAS vs. Mainframe SAS
>What was it recently? No camera phones. That didn't last long because they
>are all now camera phones. I went onsite at one place that disabled USB
&g...Re: PC SAS vs. Mainframe SAS #35
As I recall, at an SUGI opening session around 1999 or 2000, Dr. Goodnight
remarked that one firm had solved their Y2K problems by electing to install
SAP. The installation was so costly that they went bankrupt.
CONFIDENTIALITY NOTICE: This electronic message contains
information which may be legally confidential and/or privileged and
does not in any case represent a firm ENERGY COMMODITY bid or offer
relating thereto which binds the sender without an additional
express written confirmation to that effect. The information is
intended s...Re: SAS-EG Vs PC-SAS #2
Remember that EG is generating SAS code and then submitting it to a SAS server for execution; EG itself does not execute the code. If you hare seeing a 5-10 fold increase in execution time, I would suspect that your hardware/software are suboptimal for what you are doing. Another possibility may be network lag caused by having to move files on the fly.
Are your two programs (PC and EG) really the same? If they are, and you have eliminated everything within your power to control, then it may be time to start complaining vigourously to your management. A 5-10 fold increase is rediculous...
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU]On Behalf Of
Sent: Tuesday, September 05, 2006 8:40 AM
Subject: SAS-EG Vs PC-SAS
Our organization is migrating from PC-SAS to SAS-EG. It makes perfect
business sense as we are a huge organization and require too many licences.
We can have the server licence of EG for, say x, number of users. In this
way my company can also keep the count of all the licences which different
business units are using.
But the problem is that we find that EG is not performing so well in terms
of time taking to run codes as there is network lag, and the way EG handles
the data. I use some models for forecasting which takes me about an hour
each on PC, but it may take anywhere around 5-10 hours on EG. It can also be
due to all those GUI which SAS has incorporated in thi...