Date: Tue, 24 Dec 1996 21:05:00 CST
Reply-To: Undetermined origin c/o LISTSERV administrator
<owner-LISTSERV@UGA.CC.UGA.EDU>
Sender: "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From: Undetermined origin c/o LISTSERV administrator
<owner-LISTSERV@UGA.CC.UGA.EDU>
Subject: Re: SAS-L Digest - 22 Dec 1996 to 23 Dec 1996
There are 17 messages totalling 952 lines in this issue.
Topics of the day:
1. Revision Control with SAS?
2. Position open in leading MicroMarketing firm
3. Modern Regression and Classification course - Hawaii
4. Thanks for the help. I made my deadline.
5. US-IL-Chicago SAS Help Desk/Technical Support - Asen
6. US-IL-Chicago Base SAS Programmer - Asen
7. US-IL-Chicago End User Support w/ SAS, UNIX, & VM/CMS experience. - Asen
8. US-IL-Chicago Statistical SAS Programmer - Asen
9. US-IL-Chicago Senior SAS/AF Developer
10. binary data files
11. US-IL-Chicago Consultant for Validation of SAS Software - Asen
12. Connecting 2 PCS
13. Connecting 2 PCs
14. US-IL-Chicago Senior SAS Programmer with Clinical trial experience
15. Freq on unique combinations (2)
16. Menu-driven data management system
----------------------------------------------------------------------
Date: Sat, 21 Dec 1996 22:15:12 -0800
From: Karsten Self <kmself@IX.NETCOM.COM>
Subject: Re: Revision Control with SAS?
Mike,
There are several possibilities, I'll mention a couple here.
First, I'll point out the obvious -- that version control really consists of
several related functions:
- Restricting concurrent update access to objects (either on a
file/table/dataset basis, or on a row/record/line basis).
- Providing information on the most recent version of data on a system
(create date, user, notes).
- Providing rollback capabilities -- enabling restoration of prior states
of a system or file.
Some of these capabilities are easier to implement than others. Some of the
capabilities are easier to implement in some contexts (text files) than
others (binary/data files).
- I've heard some mumblings about SAS coming out with some sort of version
control/revision control product, possibly integrated with the workings of
the SQL DICTIONARY views. The fact that some objects in the views have
create times specified as dates and others as datetime makes the
implementation features of this somewhat interesting.
- Depending on your platform(s) and/or access to software, you might want to
incorporate a standard version control product into your solution. The more
current versions of RCS (a semi-standard Unix utility, also available for
Win95 and WinNT, DEC VAX/AXP) support binary files. This solution would
probably involve doubling (or more) your data storage requirement (archived
and 'checked out' data are accessed differently). I don't know whether or
which ported versions support binary file version control. Other version
control products may also be available which better suit your needs.
- SAS/SHARE, password protection, and several other features of SAS could be
used to restrict access to SAS datasets (though they wouldn't provide
rollback capabilities for your data).
- Buying a real database with rollback and versioning capabilities, and
using SAS/ACCESS to provide access to the data. If required for processing
requirements, the current view of the data could be replicated as a current
SAS dataset, other versions would be supported as views or via other mechanisms.
Good luck.
>Date: Wed, 18 Dec 1996 16:35:08 GMT
>From: Mike McDowell <mcdowell@MORPHEUS.COR.EPA.GOV>
>Subject: Revision Control with SAS?
>
>Hi all,
>
>We want to implement a revision control system for our SAS datasets -
>where we could 'check out' the latest version of a dataset, or, if
>necessary, check out an old version of a dataset. Has anyone
>implemented such a system (maybe using proc compare)? Any ideas
>on how we might go about doing this? Most (if not all) of the
>systems that I know about only do source code revision control.
>
>Thanks,
>
>Mike
>
>------------------------------------------------------------------------
>Mike McDowell OAO Corporation
>mcdowell@mail.cor.epa.gov US EPA National Health and Environmental
> Effects Research Laboratory
>(541) 754-4474 Western Ecology Division
>mcdowell@peak.org Corvallis, Oregon
----------------------------------------
Karsten Self / kmself@ix.netcom.com
What part of gestalt don't you understand?
------------------------------
Date: Sun, 22 Dec 1996 05:02:07 GMT
From: RMuellerZZ <rmuellerzz@AOL.COM>
Subject: Position open in leading MicroMarketing firm
Spectra/Market Metrics (Chicago) the leader in Consumer Packaged Goods
Micromarketing, is seeking two high energy SAS programmers/analysts.
1) Manager of Data Services
2) Marketing Analyst
For more information, email to RMuellerZZ@aol.com, or to
"Spectra_MM@msn.com" attention Dept. R (R. Mueller)
------------------------------
Date: Sat, 21 Dec 1996 20:16:22 -0800
From: Trevor Hastie <trevor@RGMILLER.STANFORD.EDU>
Subject: Modern Regression and Classification course - Hawaii
************* 1997 Course Announcement *********
MODERN REGRESSION AND CLASSIFICATION
Waikiki, Hawaii: February 17-18, 1997
*************************************************
A two-day course on widely applicable statistical methods for
modeling and prediction, featuring
Professor Trevor Hastie and Professor Robert Tibshirani
Stanford University University of Toronto
This course was offered and enthusiastically attended at five
different locations in the USA in 1996.
This two day course covers modern tools for statistical prediction and
classification. We start from square one, with a review of linear
techniques for regression and classification, and then take attendees
through a tour of:
o Flexible regression techniques
o Classification and regression trees
o Neural networks
o Projection pursuit regression
o Nearest Neighbor methods
o Learning vector quantization
o Wavelets
o Bootstrap and cross-validation
We will also illustrate software tools for implementing the methods.
Our objective is to provide attendees with the background and
knowledge necessary to apply these modern tools to solve their own
real-world problems. The course is geared for:
o Statisticians
o Financial analysts
o Industrial managers
o Medical and Quantitative researchers
o Scientists
o others interested in prediction and classification
Attendees should have an undergraduate degree in a quantitative
field, or have knowledge and experience working in such a field.
PRICE: $750 per attendee if received by January 15, 1997. Full time
registered students receive a 40% discount. Attendance is limited to
the first 60 applicants, so sign up soon! These courses fill up
quickly.
TO REGISTER: Fill in and return the form appended.
For more details on the course and the instructors:
o point your web browser to:
http://stat.stanford.edu/~trevor/mrc.html
OR send a request by
o FAX to Prof. T. Hastie at (415) 326-0854, OR
o email to trevor@stat.stanford.edu
<----------------------------- Cut Here ------------------------------->
Please print, and fill in the hard copy to return by mail or FAX
REGISTRATION FORM
Modern Regression and Classification
Monday, February 17 and Tuesday, February 18, 1997.
Hilton Hawaiian Village, Waikiki Beach, Honolulu, Hawaii.
Name ___________________________________________________
Last First Middle
Firm or Institution ______________________________________
Standard Registration ____ Student Registration ____
Mailing Address (for receipt) _________________________
__________________________________________________________
__________________________________________________________
__________________________________________________________
Country Phone FAX
__________________________________________________________
email address
__________________________________________ _______________
Credit card # (if payment by credit card) Expiration Date
(Lunch preference - tick as appropriate):
___ Vegetarian ___ Non-Vegetarian
Fee payment can be made by MONEY ORDER , PERSONAL CHECK, or CREDIT CARD
(Mastercard or Visa.) For checks and money orders: all amounts are given in
US dollar figures. Make fee payable to Prof. T. Hastie. Mail it, together
with this completed Registration Form to:
Prof. T. Hastie
538 Campus Drive
Stanford
CA 94305
USA
For payment by credit card, include credit card details above, and mail to
above address, or else FAX form to 415-326-0854
For further information, contact:
Trevor Hastie
Stanford University
Tel. or FAX: 415-326-0854
e-mail: trevor@stat.stanford.edu.
http://stat.stanford.edu/~trevor/mrc.html
REGISTRATION FEE
Standard Registration: U.S. $750 ($950 after Jan 15, 1997)
Student Registration: U.S. $450 ($530 after Jan 15, 1997)
Student registrations - include copy of student ID.
- Cancellation policy: No fee if cancellation before Jan 15, 1997.
- Cancellation fee after January 15 but before Feb 12, 1997: $100.
- Refund at discretion of organizers if cancellation after Feb 12, 1997.
- Registration fee includes course materials, coffee breaks, and lunches
- On-site Registration is possible if course is not fully booked, at late
fee.
------------------------------
Date: Mon, 16 Dec 1996 05:13:48 GMT
From: "John E. Bentley" <BentleyJ@IX.NETCOM.COM>
Subject: Thanks for the help. I made my deadline.
Thanks to all who over the past couple of days responded to my posting
about problems with a macro and with Proc DBF. I received numerous
replies by email in addition to those that were posted here.
With your help, I've finished the project and will be able to deliver
to the client tomorrow as scheduled.
John E. Bentley
Lakeview, Chicago, Illinois, USA, NorthAm, Terra
------------------------------------------------
Send Lawyers, Guns, and Money
------------------------------
Date: Wed, 18 Dec 1996 23:07:06 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago SAS Help Desk/Technical Support - Asen
Position :
Help Desk/Technical Support
Location :
North Chicago Suburb
Project Description :
Work on solving problems and addressing end user concerns. Some light
programming in SAS.
Required Skills :
SAS, UNIX, and VMS.
Required Experience :
One year of practical industry experience.
Project Duration :
Four months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail:netnews@asen.com
http://www.asen.com/
------------------------------
Date: Wed, 18 Dec 1996 23:07:20 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago Base SAS Programmer - Asen
Position :
Programmer
Location :
North Chicago Suburb
Project Description :
Work with data listings with data on paper.
Required Skills :
SAS (Base & Macro), PUT statements, Arrays, and UNIX.
Desired Skill :
Clinical trials.
Required Experience :
Two years of practical industry experience.
Project Duration :
Twelve months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail: netnews@asen.com
http://www.asen.com/
------------------------------
Date: Wed, 18 Dec 1996 23:07:23 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago End User Support w/ SAS, UNIX,
& VM/CMS experience. - Asen
Position :
Help Desk
Location :
North Chicago Suburb
Project Description :
Receive and track user contacts reporting problems with systems and
requesting assistance in using SAS based clinical systems applications in a
UNIX
and VM/CMS environment. Track and report SAS system bugs found by the SAS
programming staff to the SAS Institute. Respond to calls independently or with
other support personnel, such as the SAS development or UNIX support staff.
Light programming to generate SAS reports.
Required Skills :
SAS, UNIX, VM/CMS, pharmaceutical, and end user support.
Required Experience :
One year of experience in end user support in a help desk or training role
and some SAS, UNIX, VM/CMS, and pharmaceutical experience.
Project Duration :
Twelve months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail: netnews@asen.com
http://www.asen.com/
------------------------------
Date: Wed, 18 Dec 1996 23:07:26 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago Statistical SAS Programmer - Asen
Position :
Statistical SAS Programmer
Location :
North Chicago Suburb
Required Skill :
SAS
Desired Skill :
Clinical Trials
Required Experience :
Recent graduates with a G.P.A. of at least 3.9 out of 4.0 or at least two
years practical industry experience with a G.P.A. of at least 3.6 out of 4.0.
Project Duration :
Twelve months plus.
Rate :
Commensurate with experience and/or G.P.A. and degree.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail: netnews@asen.com
http://www.asen.com/
------------------------------
Date: Wed, 18 Dec 1996 23:07:28 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago Senior SAS/AF Developer
Position :
Programmer
Location :
North Chicago Suburb
Project Description :
Design and develop SAS-based GUI applications using SAS/AF frames on a UNIX
platform for clinical trial management and reporting.
Required Skills :
SAS/AF frames and UNIX.
Required Experience :
A minimum of fours years of SAS experience, two of which include developing
user interfaces using SAS/AF and one year of which includes SAS frame
development. Experience working independently with end users to obtain
application specifications, working in a regulatory environment, and in
validating applications.
Project Duration :
Twelve months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail: netnews@asen.com
http://www.asen.com/
------------------------------
Date: Mon, 16 Dec 1996 17:57:20 -0600
From: Erik Engstrom <eman@MINN.NET>
Subject: binary data files
Is there a SAS utility that will allow SAS to read custom binary files?
In our binary files each record is 7 bytes long and contains the
following 10 fields:
second 6 bits values range from 0 to 59
minute 6 bits values range from 0 to 59
month 4 bits values range from 1 to 12
hour 5 bits values range from 0 to 23
day 5 bits values range from 1 to 31
year 6 bits values range from 0 to 63
volume 8 bits values range from 0 to 255
occupancy 8 bits values range from 0 to 100
status 4 bits values range from -4 to +3
flag 4 bits values range from -4 to +3
It sure would be nice to get this into a SAS data set somehow any ideas?
------------------------------
Date: Wed, 18 Dec 1996 23:07:12 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago Consultant for Validation of SAS Software - Asen
Position :
Consultant
Location :
North Chicago Suburb
Project Description :
Documentation, test plans executions, and technical writing working with
SAS and UNIX.
Required Skills :
Documentation, test plans execution, and technical writing, and ability
to read and code in SAS.
Required Experience :
Two years of practical industry experience.
Project Duration :
Four months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail:netnews@asen.com
http://www.asen.com/
------------------------------
Date: Sun, 22 Dec 1996 10:54:01 -0500
From: Xlr82sas@AOL.COM
Subject: Connecting 2 PCS
Win95 contains all the networking software you need to connect two or more
PCs. I am writing this from memory, I expect it is incomplete, with errors.
I am trying to provoke comments, especially from the experts.
What you need for two PCs
For inexperienced network administrator, like me and you,:
1. Purchase two of the slower, 5 megabit ethernet cards ,make sure the
cards
are NE2000 Compatible and Plug and Play. I bought two cards from CE
for $23 dollars each. The faster 10 megabit cards are about $100
dollars each.
2. By a RJ11 phone cord, looks like phone cord with a little fatter
conectors, cord shouls be about $5.
3. Install the cards and connect with the RJ11 cord.
4. You may need to set the addresses for each card, see doc with
ethernet cars.
5. When you restart each computer, I think Win95 will automatically
add the cards. If it does not, go to control panel and add the
cards.
Also make sure you have netbui protocol installed.
6. You will also need to assign a 'name' for each system. I used
SAS001 for the primary system, the system that contains the
names of all other systems,. I named the second bare bones
system SAS002. I think you need to use the Network ICON in
control panel to do this. It is nicely menu driven.
7. Once you have done the above. Restart the barebones system SAS002.
You should see a new ICON in window explorer, I think it is
Network Files??, you should be able to execute, browse and
edit files on SAS001.
8. To access file on SAS001 from SAS002 just precede the
direcory name withj SAS001/, I think.
9. I have been told that you do not even need to have Win95 on SAS002.
I am not a client/server expert, but all this very expensive server
technology
seems like overkill to me.
Suppose you have a large read only or read now change with batch later
database. Also suppose the network is a small department level net,
less than 6 users. Why not just connect the dept systems using Win95 netbui.
Put SAS on SAS001 and let the users execute SAS on SAS001 with the
dispaly and output directed to their PCs. I am pretty sure this is possible
in a Unix client/server enviironment. No sure it is possible under Win95.
Not sure of the lincense issues.
CompuCraft Inc
Roger DeAngelis
------------------------------
Date: Sun, 22 Dec 1996 13:39:00 EST
From: "Bassett Consulting Services, Inc." <0002395748@MCIMAIL.COM>
Subject: Re: Connecting 2 PCs
-----------------------------------------------------------------------
CONTENT: Re: Connecting 2 PCs
SUMMARY: warning: slightly off-list topic
REL/PLTF: 6.11/WfW, Win 95
NAME: Michael L. Davis
X.400: C=US ; A=MCI ; S=DAVIS ; DDA=ID=2395748
INTERNET: Bassett.Consulting@mcimail.com
AFFILIATION: Bassett Consulting Services, Inc.
P-ADDR: 10 Pleasant Drive, North Haven, CT 06473
PHONE: (203) 562-0640
VOICE MAIL: (203) 580-0170 (toll-free in Conn.)
FAX: (203) 498-1414
-----------------------------------------------------------------------
My friend and prolific SAS-L contributor, Roger DeAngelis
<Xlr82sas@aol.com> posted a message earlier today about establish a
network connection between two PCs. While the subject is slightly
off-list topic, I believe that his information and what I will supply
will be of interest to many SAS-Lers. However, if I am wrong in your
case, please accept my apologies for consuming the bandwidth
unnecessarily.
First, both Windows for Workgroups and Windows 95 both have the
software required to set up a peer-to-peer network. In fact, this
message is being composed on a two PC network where a Windows 95 and
a Windows for Workgroups machine are networked via ethernet.
When an inexperienced person purchases ethernet cards, they ought to
give thought to buying the cards from a vendor who can also supply
technical support, even if it means paying a premium for the card.
If one decides ignor this advice and purchase adapters from the least
expense source, budget $35 for support from Microsoft when the
inevitable problems emerge.
Also, my experience has been that combo cards (that accept RJ-45,
coaxial, and "DB" connectors) cost little more and will increase your
flexibility. Another tip is to purchase the same brand cards for all
nodes of your network to avoid incompatibility problems.
Last, I use PnP (plug and play) adapters. Even if your motherboard
does not support PnP, it guarantees that you can alter your adapter's
setting through software rather than first turning to the "can opener".
When Roger said:
> 2. Buy a RJ11 phone cord, looks like a phone cord with a little
> fatter connectors, cord should be about $5.
I believe that the cord he referred to is designated as RJ-45 and has
eight conductors. RJ-11 cords, commonly used with non-digital
telephones have four conductors.
Roger advises connecting the PCs with an RJ-45 cable. My advice on
the cabling is to use a coaxial cable (also known as "Thinnet") with
BNC connectors pre-installed, along with two BNC T-adapters and
terminating 75 ohm resistors on the open ends of the T-adapters.
My rationale is that other wiring set-ups usually require one to buy
and set up a hub. With the set-up that I have described, no hub is
necessary. Also, installing BNC connectors correctly can be tricky.
If possible, obtain pre-made cables.
Regarding networking protocols, I am using IPX/SPX with Netbios
instead of Netbui in order to support Symantec's PC-Anywhere. I use
PC-Anywhere to allow me to remotely operate the other computer on the
network.
One problem that I had when I started was that I had not provided for
enough file handles. The problem went away I bumped up the value of
FILES= in my CONFIG.SYS files.
Roger also wrote:
> Why not just connect the dept systems using Win95 netbui. Put SAS
> on SAS001 and let the users execute SAS on SAS001 with the display
> and output directed to their PCs. I am pretty sure this is possible
> in a Unix client/server environment. Not sure it is possible under
> Win95. Not sure of the license issues.
The reason why you can do this under Unix client/server is because
the host is executing the SAS System and the clients are essentially
terminals. To emulate this situation under Windows, you have to use
remote control software, such as PC-Anywhere.
I don't speak for SI and their licensing requirements. However, I
suspect that an additional license is required for each computer on
the network that is executing SI software. I don't know if using
SAS on a single CPU via remote control software reduces the number
of computer licenses that must be obtained.
Happy Holidays to my fellow SAS-lers.
------------------------------
Date: Wed, 18 Dec 1996 23:55:03 GMT
From: Asen Computer Associates <netnews@ASEN.COM>
Subject: US-IL-Chicago Senior SAS Programmer with Clinical trial experience
Position :
Senior Programmer/Analyst
Location :
North Chicago Suburb
Project Description :
Analysis and programming using SAS.
Required Skills :
SAS (Macro, Graph, Stat, and Base).
Required Experience :
One year of programming tables and listings with various and sundry SAS
modules.
Project Duration :
Four months plus.
Rate :
Commensurate with experience.
Now in our twenty-second year, Asen Computer Associates is a leading Chicago
area computer consulting firm. To meet the demand of providing specialists to
Fortune 500 companies, we use a combination of project-basis and full-time
consultants. If you are interested in the above position, please e-mail
(preferably in ASCII) or fax your resume and any other pertinent information
to:
Denise Peters
Asen Computer Associates, Inc.
900 North National Parkway, Suite 155
Schaumburg, IL 60173
Voice: +1 847 995 1300
Fax : +1 847 995 1305
E-mail: netnews@asen.com
http://www.asen.com/
------------------------------
Date: Sun, 22 Dec 1996 09:02:09 -0500
From: Ron Coleman <rcoleman@WORLDNET.ATT.NET>
Subject: Re: Freq on unique combinations
SheridanY wrote:
>
> I have a table created through proc sql which has seven fields that are
> populated with values 1-5. There fields that have missing values as well.
> If I want to look at the distribution of these values in unique
> combinations, what is the easiest way? An example would be:
>
> Field A B C D E F G
> Case 1 1 2 3 4 4 5 1
> Case 2 5 3 3 2 2 2 1
> Case 3 5 4 - 2 1 1 1
>
> So the question is how many A and B combinations versus A, B, C
> combinations versus A,B,C,D versus A,C combinations exist regardless of
> number of cases. I just need to figure out the syntax prior to running
> proc freq on the different combinations.
>
> Any suggestions would be helpful.
>
> Thanks.
>
> Sheridan Young
Sheridan,
My favorite method is to create a character variable that shows the
unique cases and then do a frequency on that.
data unique(keep=unique);
set whatever;
length unique $7;
array values(7) a b c d e f g;
do i = 1 to dim(values);
substr(unique,i,1) = put(values(i),1.);
end;
run;
proc freq data=unique;
tables unique;
run;
... would produce something like:
Cumulative Cumulative
UNIQUE Frequency Percent Frequency Percent
-----------------------------------------------------
1115121 1 10.0 1 10.0
1121255 1 10.0 2 20.0
1453145 1 10.0 3 30.0
2222244 1 10.0 4 40.0
2325512 1 10.0 5 50.0
2415442 1 10.0 6 60.0
2421523 1 10.0 7 70.0
2544153 1 10.0 8 80.0
3254442 1 10.0 9 90.0
3414151 1 10.0 10 100.0
--
Ron Coleman
Links Analytical, Inc. Linking your data to your business!
3545-1 St. Johns Bluff Rd. Suite 300
Jacksonville FL 32224
rcoleman@worldnet.att.net
------------------------------
Date: Sun, 22 Dec 1996 12:49:17 -0800
From: "T. Grichuhin" <ted@U.WASHINGTON.EDU>
Subject: Re: Freq on unique combinations
You could also create an index containing A-G as keys, then do a first.G
in a data statement, then do the frequencies on the variable combinations
you are interested in.
On Sun, 22 Dec 1996, Ron Coleman wrote:
> SheridanY wrote:
> >
> > I have a table created through proc sql which has seven fields that are
> > populated with values 1-5. There fields that have missing values as well.
> > If I want to look at the distribution of these values in unique
> > combinations, what is the easiest way? An example would be:
> >
> > Field A B C D E F G
> > Case 1 1 2 3 4 4 5 1
> > Case 2 5 3 3 2 2 2 1
> > Case 3 5 4 - 2 1 1 1
> >
> > So the question is how many A and B combinations versus A, B, C
> > combinations versus A,B,C,D versus A,C combinations exist regardless of
> > number of cases. I just need to figure out the syntax prior to running
> > proc freq on the different combinations.
> >
> > Any suggestions would be helpful.
> >
> > Thanks.
> >
> > Sheridan Young
>
> Sheridan,
>
> My favorite method is to create a character variable that shows the
> unique cases and then do a frequency on that.
>
> data unique(keep=unique);
> set whatever;
> length unique $7;
>
> array values(7) a b c d e f g;
>
> do i = 1 to dim(values);
> substr(unique,i,1) = put(values(i),1.);
> end;
> run;
> proc freq data=unique;
> tables unique;
> run;
>
> ... would produce something like:
>
> Cumulative Cumulative
> UNIQUE Frequency Percent Frequency Percent
> -----------------------------------------------------
> 1115121 1 10.0 1 10.0
> 1121255 1 10.0 2 20.0
> 1453145 1 10.0 3 30.0
> 2222244 1 10.0 4 40.0
> 2325512 1 10.0 5 50.0
> 2415442 1 10.0 6 60.0
> 2421523 1 10.0 7 70.0
> 2544153 1 10.0 8 80.0
> 3254442 1 10.0 9 90.0
> 3414151 1 10.0 10 100.0
>
>
> --
> Ron Coleman
> Links Analytical, Inc. Linking your data to your business!
> 3545-1 St. Johns Bluff Rd. Suite 300
> Jacksonville FL 32224
> rcoleman@worldnet.att.net
>
------------------------------
Date: Sun, 22 Dec 1996 08:47:55 -0500
From: Ron Coleman <rcoleman@WORLDNET.ATT.NET>
Subject: Re: Menu-driven data management system
Junming Zhu wrote:
>
> First, I would like to thank for all who offering help before. I have
> another question: Is it possible to design a menu-driven data management
> system using SAS? I have many tables, my objective is to design a data
> retrieving system so that others can select the tables, fields and
> rows in different tables from the menu. If it is feasible, can anyone
> suggest any references so that I can learn? Thanks in advance.
>
> Zhu Junming
A real quick solution is to use SAS/ASSIST. It covers about 70-80% of
SAS functionality.
--
Ron Coleman
Links Analytical, Inc. Linking your data to your business!
3545-1 St. Johns Bluff Rd. Suite 300
Jacksonville FL 32224
rcoleman@worldnet.att.net
------------------------------
End of SAS-L Digest - 22 Dec 1996 to 23 Dec 1996
************************************************
