Assessing the relative strength of driver variables in a multivariate logistic regression

2010-02-04T10:07:07Z

Hello,

Which method do you use to assess the relative strength of each driver in a multivariate logistic regression ? This is very useful to show this on a pie chart to a non modeller audience.

Kind regards,

Toloc

How do I edit questions on run;submit?

2009-11-13T19:43:18Z

I have entered an answer to a question, which was edited by some helpful other user. I would like to do an additional edit, since a problem remains. How do I do that?

I also tried to send an email to info@runsubmit.com, but it bounced.

Does ANYONE use SAS Servers or SAS BI and not hate it?

2010-01-09T01:53:50Z

Are there any active, large installations of the SAS Business Intelligence framework? Is anyone actively using all of the capabilities of the Metadata Server? Does anyone actively use SAS Information Maps? Does anyone use anything more than Base/Stat/Graph/IML/Access, etc.

Is SAS' huge attempt at "Enterprise Class" shared data and analytics just a 10 year waste of time? Or are there real success stories?

I've only ever heard of companies and customers "beginning" a SAS BI deployment and companies/customers "tearing out" a SAS BI Deployment.

Some encouraging success stories would be nice. Anyone?

How do I change the picture that appears beside my name?

2009-10-18T19:15:36Z

I want to change the picture that appears beside my name. How do I do that?

How do you determine the size of sas dataset on UNIX?

2009-12-18T15:41:31Z

Does anyone know of any SAS functions / techniques for obtaining the SIZE of files / datasets in UNIX directories? (eg in MB or GB).

How do I programmatically create New Catalogs?

2010-01-15T14:22:46Z

I suspect I missing something really obvious here, but is there a way to programatically create a new catalog?

I've had a look at PROC CATALOG but can't see anything which would create a new one.

(I did try to tag this as Catalogs/PROC Catalog, but apparently these are new tags and I can't create them)

Finding out the name and path of the program being run

2009-11-06T11:10:55Z

How can I determine the name and path of the program I am currently running? (In Windows, sas 9.1.3 - but techniques for other versions / environments would also be good to know..)

SAS Coding Standards

2009-12-16T16:36:13Z

What are some useful and necessary coding standards that most, if not all, SAS programmers should adhere to?

One suggestion/coding standard or principle per Answer. Vote up the answers you agree with.

Disabling SAS modules?

2010-01-09T23:54:00Z

I have SAS/ACCESS licensed (among other modules) but want to disable it in order to simulate someone else's environment? It would also be helpful for answering some of the questions on this site. Is anyone aware of a way to do this?

Maximum size of a SAS script

2010-01-12T09:13:50Z

Is there a maximum size of a SAS SCRIPT? If so, how does this differ (if at all) between SAS 8 and 9?

Identify first and last value by date collected per record

2009-12-04T02:53:20Z

Here is how my data are set up:

where date# is the date the lab result# was collected. I have tried to check if the dates are in asending order but cannot be certain (the real data can have up to 10 dates with 10 corresponding lab results). Where . are missing data. Where DATE# are dates mm/dd/yyyy, RESULT# are numeric, and ID is a string.

ID DATE1 RESULT1 DATE2 RESULT2 DATE3 RESULT3 DATE4 RESULT4

1   1/1/2000   200   3/1/2000   250   8/1/2000   100   9/1/2000   400  

2   .    .      .    .   .    .   .   .

3   2/1/2000   225   9/1/2000   100   .   .   .   .

4   12/1/2000   300   .   .   .   .   .   .

QUESTION 1: How do I identify/tag the earliest and oldest lab result per ID?

These here is the output I need:

ID FIRST LAST

1   200   400

2   .     .

3   225   100

4   300   300

QUESTION2: any way to make the LAST output for ID 4 set to missing since there is only one result for this record?

Is a Hash object faster than an Array?

2009-10-21T01:51:23Z

Could anyone explain why an array will always be faster than a hash object described in SAS Prog3 course note?

FileNotFoundException when using Enterprise Guide automation server with C#

2009-12-07T22:49:10Z

Any idea why I get this exception when using C# to start up the Enterprise Guide automation server?

Code:

SAS.EG.Scripting.Application enterpriseGuide = new SAS.EG.Scripting.Application();
SAS.EG.Scripting.ISASEGProject project = enterpriseGuide.New();

throws FileNotFoundException

with Message "C:\SAIC Development\Main\SAS\working\alerts\add-in\AlertManager\bin\Debug\SEGuide.exe"

and Stack Trace

at System.Diagnostics.FileVersionInfo.GetVersionInfo(String fileName)
   at SAS.Shared.SharedEnvironment.get_ApplicationVersion()
   at SAS.EG.ProjectElements.Element.SetModified()
   at SAS.EG.ProjectElements.Element.SetLabel(String strLabel)
   at SAS.EG.ProjectElements.ProjectCollection..ctor(IMetadataService metadataService)
   at SAS.EG.ProjectElements.ProjectCollection..ctor()
   at SAS.EG.Scripting.Project..ctor()
   at SAS.EG.Scripting.Application.New()
   at AlertManager.AlertDesignerForm.CreateNew() in C:\SAIC Development\Main\SAS\working\alerts\add-in\AlertManager\AlertDesignerForm.cs:line 50
   at AlertManager.AlertDesignerForm.newToolStripMenuItem_Click(Object sender, EventArgs e) in C:\SAIC Development\Main\SAS\working\alerts\add-in\AlertManager\AlertDesignerForm.cs:line 128
   at System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
   at System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
   at System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
   at System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
   at System.Windows.Forms.ToolStripItem.FireEventInteractive(EventArgs e, ToolStripItemEventType met)
   at System.Windows.Forms.ToolStripItem.FireEvent(EventArgs e, ToolStripItemEventType met)
   at System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
   at System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
   at System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
   at System.Windows.Forms.Control.WndProc(Message& m)
   at System.Windows.Forms.ScrollableControl.WndProc(Message& m)
   at System.Windows.Forms.ToolStrip.WndProc(Message& m)
   at System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
   at System.Windows.Forms.Control.ControlNativeWindow.OnMessage(Message& m)
   at System.Windows.Forms.Control.ControlNativeWindow.WndProc(Message& m)
   at System.Windows.Forms.NativeWindow.DebuggableCallback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
   at System.Windows.Forms.UnsafeNativeMethods.DispatchMessageW(MSG& msg)
   at System.Windows.Forms.Application.ComponentManager.System.Windows.Forms.UnsafeNativeMethods.IMsoComponentManager.FPushMessageLoop(Int32 dwComponentID, Int32 reason, Int32 pvLoopData)
   at System.Windows.Forms.Application.ThreadContext.RunMessageLoopInner(Int32 reason, ApplicationContext context)
   at System.Windows.Forms.Application.ThreadContext.RunMessageLoop(Int32 reason, ApplicationContext context)
   at System.Windows.Forms.Application.Run(Form mainForm)
   at AlertManager.Program.Main() in C:\SAIC Development\Main\SAS\working\alerts\add-in\AlertManager\Program.cs:line 18
   at System.AppDomain._nExecuteAssembly(Assembly assembly, String[] args)
   at System.AppDomain.ExecuteAssembly(String assemblyFile, Evidence assemblySecurity, String[] args)
   at Microsoft.VisualStudio.HostingProcess.HostProc.RunUsersAssembly()
   at System.Threading.ThreadHelper.ThreadStart_Context(Object state)
   at System.Threading.ExecutionContext.Run(ExecutionContext executionContext, ContextCallback callback, Object state)
   at System.Threading.ThreadHelper.ThreadStart()

Any help would be appreciated.

Finding the # of days between two dates?

2009-10-30T11:11:36Z

I have data like:

 patid             startdate                    endday 
01S1003            9/30/2008                 3/16/2009

How do I find the number of days bewteen startdate and endday?

SAS Support for Enterprise Guide Add-In using VS2008?

2009-11-04T16:36:07Z

Does SAS have any examples or templates/wizards that are usable in Visual Studio 2008? I'm trying to explore building custom tasks and I have EGuide installed along with MS-Addin as well. I can open the example projects in VS2008 but I'm missing references. Any ideas? Also, the EG Wizard and Project types aren't installed properly (I manually copied based on my review of the Deploy.vbs file.

Any help or hints would be appreciated...

How to suppress the display of output dataset nodes in EG?

2010-01-07T02:43:23Z

In Enterprise Guide, any output datasets (created in WORK or other LIBNAME) are, by default, displayed as output nodes emitting from the program or code node that generates them.

I have a relatively complex process that generates (by necessity) a TON of intermediate work datasets. I'd rather not kill them all as a part of the process ( I guess that's one solution).

Is there a way (through options or registry keys or devil magic) to suppress the display of these gajillion output datasets?

Copying files from RWORK to WORK using EG

2010-01-04T17:48:09Z

Can anyone advise how to perform a 'proc download' equivalent using EG?

ie copy files from the WORK library held on the remote server (SASUnix) to the local WORK library?

There must be a workaround!!

Theta Symbol in a SAS Label?

2010-01-04T19:58:58Z

How do I make the label for a variable display a Greek character.

So I need:

label Theta = <somehex_expression?>;

I'm trying to label the variable named Theta with the upper case Greek theta symbol. thoughts?

Error with ODS and Proc Template

2009-12-29T22:56:31Z

I am trying to create a custom template within a program run via Enterprise Guide. However, no matter what I do, I keep getting an error:

ERROR: Template 'MyTemplate' was unable to write to template store!

Is this a permissions issue of some kind?

Replacing missing values by the medians within each sub group

2009-12-19T15:57:58Z

Hello everyone,

I am trying to replace missing values with the corresponding median, bearing in mind that this Median must varies across a class variable (CountryCode)

so I have something like

CountryCode AGE US 12 US 10 US . FR 11 FR . NL 10

etc...

in the above example I am trying to replace the missing in Age by the medians AGE within each Country (US, FR, NL)

Have tried a coupld of sulutions that will relace with the median across the whole sample but nothing that actually take into account the class variable CountryCode)

Thanking you in advance,

Proper SAS format for a difference in time XXXX

2009-10-18T15:52:39Z

Can anyone suggest the best SAS format for a difference in time?

From what I can tell the TIMEw. & HHMMw. & MMSSw. are not appropriate b/c they report time from midnight. I just want to take the number of seconds between to time points and express it in hours, minutes, seconds as appropriate.

Code or process to assign value labels to specific numerical values in a dataset

2009-12-02T22:12:52Z

I am still struggling a bit with how SAS handles value labels for specific numbers in a variable.

How do you code value labels within a variable/column for a SAS dataset?

An example is LevelofCare variable: 5=education 6=medication assist 10=outpatient 20=intensive outpatient 30=residential

In Stata I would do such: label define LOCSE 5 "Education" 6 "Methadone" 10 "Outpatient" 20 "Intensive Outpatient" 30 "Residential" 40 "Detox" label values LevelofCare LOCSE

The result is I would then have value labels assigned to that Stata dataset permanently.

Is there such a process or command in SAS?

I have tried and know you can do a proc format statement like:

proc format;

value $CareLevellb
'0'='Assess only'
'00'='Assess only'
'05'='Education'
'10'='Outpatient'
'20'='Intensive outpatient'
'25'='Day tx'
'30'='Residential';
attrib LevelCare label="Level of care received";
format LevelCare $CareLevellb.;
RUN;

But I believe this is only temporary. What am I doing incorrectly? I did try variations running the above say under a data command such as

DATA Rfpdata.Test2008; SET Rfpdata.Tb2008_nodup proc format; ..............

Any suggestions? Will/how does SAS incorporate variable value labels in a permanent dataset once you run a program.

Thank-you ahead of time.

hash table 'scalar' error

2009-12-15T16:30:02Z

Does anyone have any idea why I am getting the following error?

ERROR: Scalar cannot be converted to object of type hash.

the hash table has three columns and one row. I am loading it in a data step that has no set statement - see below:

data dataset1;
 imp='TOTAL';
 var1=3;
 var2=4;
run;

data test;
 call missing(var1,var2);
 declare hash imp(dataset:"dataset1 ");
 imp.defineKey('IMP');
 imp.defineData("VAR1","VAR2");
 imp.defineDone();
 imp='TOTAL';
 if imp.find() = 0 then PVIF= sum(var1, 2);
run;

can anyone advise?

way to use text editor on runsubmit

2009-12-04T08:36:35Z

Could you please tell me the writing conventions to use on runsubmit ? To answer or ask questions it is not always possible to write special characters (like the underscore), I've got also some time parts of SAS code which doesn't appear.

Kind Regards
Toloc

OLEDB Excel Connection in Read-Only Mode?

2009-12-07T18:12:39Z

I'm having trouble getting OLEDB for Excel to work in read-only mode. The document is usually accessed by another user, and I'd like to import its contents to SAS without needing to tell other users to close down Excel first.

This is what I have right now, which is NOT opening the document in readonly mode.

libname oledbxls oledb ACCESS=READONLY provider="Microsoft.Jet.OLEDB.4.0"
properties=("data source"="&xlsfile." "Mode"="Read")
provider_string="Excel 8.0;ReadOnly=True;HDR=no;IMEX=1;"
;

This is the error I get:

ERROR: Error trying to establish connection: 
Unable to Initialize: The Microsoft
Jet database engine cannot open the file ''.  
It is already opened exclusively by another user,
or you need permission to view its data.

ERROR: Error in the LIBNAME statement.

Everything works fine unless someone has the document opened in Excel. Any advice is greatly appreciated.

Placeholder Rows/Values from XML libname

2009-11-26T15:52:45Z

I have some XML data which I'm trying to read into SAS 9.1.3 through the XML libname engine. Said XML data looks a lot like this (some may recognise that's it's output from PROC METADATA):

<?xml version="1.0" encoding="windows-1252"?>
<GetMetadataObjects>
  <Reposid>A0000001.AREPOSID</Reposid>
  <Type>Person</Type>
  <Objects>
    <Person Id="AREPOSID.AA000001" Name="User The 1st">
      <Logins/>
    </Person>
    <Person Id="AREPOSID.AA000002" Name="User The 2nd">
      <Logins>
        <Login Id="AREPOSID.AV000001" UserID="AUserID">
          <Domain>
            <AuthenticationDomain Id="AREPOSID.AD000001" Name="DefaultAuth"/>
          </Domain>
        </Login>
        <Login Id="AREPOSID.AV000002" UserID="AlsoUserID">
          <Domain>
            <AuthenticationDomain Id="AREPOSID.AD000002" Name="DummyAuth"/>
          </Domain>
        </Login>
      </Logins>
    </Person>
    <Person Id="AREPOSID.AA000003" Name="Person The 3rd">
      <Logins/>
    </Person>
  </Objects>
  <NS>SAS</NS>
  <Flags>260</Flags>
  <Options>
    <Templates>
      <Person Id="" Name="">
        <Logins Id=""/>
      </Person>
      <Logins>
        <Login/>
      </Logins>
      <Login Id="" Userid="">
        <Domain/>
      </Login>
      <Domain> 
        <AuthenticationDomain/>
      </Domain>
      <AuthenticationDomain Name=""/>
    </Templates>
  </Options>
</GetMetadataObjects>

The above XML is associated with a file reference called xml_out, and I'm using the following code to read it into a SAS dataset:

filename xml_map temp;
data _NULL_;
  file xml_map;
  put 
    '<?xml version="1.0" encoding="windows-1252"?>'
    /   '<SXLEMAP xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" name="TreeMap" version="1.2" xsi:noNamespaceSchemaLocation="http://www.sas.com/xml/schema/sxle12.xsd">'
    /   '    <TABLE name="user_logins">'
    /   '       <TABLE-PATH syntax="XPath">/GetMetadataObjects/Objects/Person/Logins/Login</TABLE-PATH>'
    /   '        <COLUMN name="UserMetaId">'
    /   '            <PATH syntax="XPath">/GetMetadataObjects/Objects/Person/@Id</PATH>'
    /   '            <TYPE>character</TYPE>'
    /   '            <DATATYPE>string</DATATYPE>'
    /   '            <LENGTH>200</LENGTH>'
    /   '        </COLUMN>'
    /   '        <COLUMN name="UserMetaName">'
    /   '            <PATH syntax="XPath">/GetMetadataObjects/Objects/Person/@Name</PATH>'
    /   '            <TYPE>character</TYPE>'
    /   '            <DATATYPE>string</DATATYPE>'
    /   '            <LENGTH>200</LENGTH>'   
    /   '        </COLUMN>'
    /   '        <COLUMN name="LoginsGroup">'
    /   '            <PATH syntax="XPath">/GetMetadataObjects/Objects/Person/Logins/@Id</PATH>'
    /   '            <TYPE>character</TYPE>'   
    /   '            <DATATYPE>string</DATATYPE>'   
    /   '            <LENGTH>200</LENGTH>'
    /   '        </COLUMN>' 
    /   '        <COLUMN name="LoginDomain">'   
    /   '            <PATH syntax="XPath">/GetMetadataObjects/Objects/Person/Logins/Login/Domain/AuthenticationDomain/@Name</PATH>'   
    /   '            <TYPE>character</TYPE>'
    /   '            <DATATYPE>string</DATATYPE>'
    /   '            <LENGTH>200</LENGTH>'
    /   '        </COLUMN>'   
    /   '        <COLUMN name="LoginId">'
    /   '            <PATH syntax="XPath">/GetMetadataObjects/Objects/Person/Logins/Login/@Id</PATH>'
    /   '            <TYPE>character</TYPE>'   
    /   '            <DATATYPE>string</DATATYPE>'   
    /   '            <LENGTH>200</LENGTH>'
    /   '        </COLUMN>'   
    /   '    </TABLE>'
    /   '</SXLEMAP>'
  ;
run;
libname xml_out xml xmlfileref=xml_out xmlmap=xml_map;
data logins_details_test;
  set xml_out.user_logins;
run;

The output data from this procedure looks like this:

UserMetaId         UserMetaName  LoginDomain  LoginId
---------------------------------------------------------------
AREPOSID.AA000002  User The 2nd  DefaultAuth  AREPOSID.AV000001
                                 DummyAuth    AREPOSID.AV000002

Can anybody suggest what I'd need to change in either the xml map, or in the libname statement so that I get either 1 row per login found, or a single row with no login details if no login details are visible? i.e.

UserMetaId         MetaName      LoginDomain  LoginId
---------------------------------------------------------------
AREPOSID.AA000001  User The 1st
AREPOSID.AA000002  User The 2nd  DefaultAuth  AREPOSID.AV000001
AREPOSID.AA000002  User The 2nd  DummyAuth    AREPOSID.AV000002
AREPOSID.AA000003  User The 3rd

Proc Transpose All Variables?

2009-11-25T21:44:04Z

I'm writing a macro that takes in a dataset with an unknown number of columns (both character and numeric), but it is known that 1 column is called obsId.

I'd like to proc transpose the contents of the entire table by ObsId, without knowing the names of the other columns. Is this doable without doing something crazy like looking it up in dictionary.columns?

Sample Dataset

ObsId Name Age Height
1     John 15  170 
2     Joe  34  160

Desired Output

ObsId _name_ col1
1     Name   John
1     Age    15
1     Height 170
2     Name   Joe
2     Age    34
2     Height 160

Equivalent Code

proc transpose data=dataset;
by ObsId;
var Name Age Height;
run;

Again, I'd like to be able to do this with any generic dataset which has a column named ObsId and an unspecified number of other columns.

Any input would be appreciated.

Tree visualization when using proc arboretum in SAS /STAT

2009-11-24T08:44:59Z

Hi,

Does anybody know how to display (in HTML format) a tree resulting from a proc arboretum in SAS STAT ?

IS it possible? if yes how?

Thanks in advance for your help JCP

GLMSELECT - Tweak for logistic regressions

2009-11-23T15:59:47Z

Hey Folks,

At the moment i am using GLMselect as a first stage investogation on potential valid and stable regressors (Out of Sample validation and ASE being the key features I want to leverage here). My dependent is a 1/0 binomial but I was hoping I could get a syntax or use another PROC (perhaps Transreg?) to transform it into the logit input - my problem is that I am using a cross section and i am not too sure about the best syntax or Proc to use - your thoughs and suggestions are much appreciated !

Unduplicating records within a SAS dataset

2009-10-20T17:53:51Z

I am new to SAS but have background programming in C++, Stata, SPSS, even JMP.

Our datasets contain multiple duplicates and I need to unduplicate the SAS records/rows within the files using multiple variables, (var1, var2, var3, var4, var5). A unique record is basically with var 1 to var 5 are a multiple key combination to indicate a unique episode of care.

In Stata I would do something like: duplicates drop (var1 var2 var3 var4 var5), force

This would drop multiple records/rows that had the same combination value of var1 to var5

Thanks for any help.

Top Questions - RunSubmit.com