Re: nested logit #2

Lee:I was writting for you.
Do you receb my comments? .
I was sending one idea for you problem.
I was working about that=2C and you said of residuals=2C I believe that is =
decision probability. I'm  thinking about that. How can  I graphic of decis=
ion probabillity?
Sandra Vergara> Date: Thu=2C 23 Oct 2008 16:13:46 -0400> From: lpanas@brand=> To: sa_vergara@HOTMAIL.COM> CC: SAS-L@LISTSERV.UGA.EDU> Subject: R=
e: nested logit> > Sandra=2C Thanks for your response. I also get red inste=
ad of blue on the utility statement. I don't think that means anything. Som=
etimes that happens with new procedures. > > Lee> > ----- Original Message =
-----> From: "sandra vergara cardozo" <sa_vergara@HOTMAIL.COM>> To: SAS-L@L=
ISTSERV.UGA.EDU> Sent: Thursday=2C October 23=2C 2008 3:11:15 PM GMT -05:00=
 US/Canada Eastern> Subject: Re: nested logit> > Lee=2C When you running of=
 program MDC=2C of the words utility appear red? or blue? In my case=2C app=
ear red? It=B4s wrong? Thanks> Date: Mon=2C 20 Oct 2008 19:42:46 -0200> Fro=
m: scardozo@ESALQ.USP.BR> Subject: [Fwd: nested logit.]> To: SAS-L@LISTSERV=
..UGA.EDU> > Lee:> > You need writte:> First:> need writting of same name os=
 variables (5):> personid =3Dtacl1> decision =3Dtacl2> topmode=3Dtacl3> mid=
mode =3D tacl4> mode =3D tacl5> > What is the variable Choice? male depress=
ion> > > data newdata1(keep=3Dpid decision mode tacl)=3B> set owl1=3B> arra=
y tvec{5} tacl1 - tacl5=3B> retain pid 0=3B> pid + 1=3B> do i =3D 1 to 5=3B=
> mode =3D i=3B> tacl =3D tvec{i}=3B> decision =3D ( choice =3D i )=3B> out=
put=3B> end=3B> run=3B> > proc mdc data=3Dnewdata1=3B> model decision =3D t=
acl / type=3Dnlogit choice=3D(mode 1 2 3 4 5 )> id pid=3B> utility u(1=2C) =
=3D tacl=3B> nest level(1) =3D (1 2 @ 1=2C 3 4 @ 2=2C 5 @ 3)=2C> level(2) =
=3D (1 2 @ 1=2C 3 @ 2)=3B> output out=3D owl1GL p=3D pdec1GL=3B> run=3B> > =
> data saida1=3B merge owl1GL=3B> e =3D tacln - pdec1GL=3B> proc print data=
=3Dsaida1=3B> run=3B> > And you can export of residuals for excel.> > I hop=
e help you.> > Sandra> > --------------------------- Mensagem Original ----=
------------------------> Assunto: nested logit.> De: "Lee Panas" <lpanas@B=
RANDEIS.EDU>> Data: Seg=2C Outubro 20=2C 2008 14:41> Para: SAS-L@LISTSERV.U=
GA.EDU> -------------------------------------------------------------------=
-------> > Hi=2C I am trying to use PROC MDC for a nested logit. My study i=
nvolves a> group of people with bipolar disorder. I am trying to predict dr=
ug> prescribed based on patient characteristics (age=2C gender=2Cetc) and c=
linical> status (depression=2C mania=2C etc.). All independent variables ar=
e individual> specific.> > At the bottom level=2C there are 5 choices for d=
rugs: 1=2C 2=2C 3=2C 4=2C 5. The> drugs are nested as follows: At mid level=
=2C there are three drug types: A> (1=2C2)=2C B (3=2C4)=2C C (5). At the to=
p level=2C there are two classes: X (A=2CB)=2C> Y (C).> > Here is part of m=
y data set:> > personid decision topmode midmode mode male depression> 1 0 =
1 1 1 1 0> 1 1 1 1 2 1 0> 1 0 1 2 3 1 0> 1 0 1 2 4 1 0> 1 0 2 3 5 1 0> 2 1 =
1 1 1 0 1> 2 0 1 1 2 0 1> 2 0 1 2 3 0 1> 2 0 1 2 4 0 1> 2 0 2 3 5 0 1> > My=
 problem is that it seems as if each indep variable can only occur once at>=
 each level. If they appear more than once=2C then SAS gives a parameter> e=
stimate but no standard error on the second occurrence. Is there any way> t=
o get SAS to print standard errors each time? Here is an example of my> SAS=
 code:> > proc mdc> model decision=3Dmale depression/type=3Dnlogit choice=
=3D(mode 1 2 3 4 5)=3B> id personid=3B> utility u(1=2C 1 2) =3D male depres=
sion> u(1=2C 3 4) =3D male depression> u(2=2C 1 2) =3D male depression> =3B=
> > nest level(1)=3D (1 2 @ 1=2C 3 4 @ 2=2C 5 @ 3)> level(2)=3D (1 2 @ 1=2C=
 3 @ 2)=3B> > Output:> > Parameter Estimate std err> > male_L1G1 -1.41 miss=
ing> depression_L1G1 2.61 3.89> male_L1G2 -1.31 1.56> depression_L1G2 2.48 =
missing> male_L2G1 2.02 5.94> depression_L2G1 -3.96 11.19> > Is there any w=
ay to get SAS to compute standard errors in the cases above> where it is mi=
ssing?> > Thanks=2C> > Lee> > > Sandra Vergara Cardozo> Doutoranda em Estat=
=EDstica> ESALQ/USP> Tel: (19)3429-4144> Ramal:233/235> > Express yourself =
instantly with MSN Messenger! MSN Messenger > _____________________________=
____________________________________> Express yourself instantly with MSN M=
essenger! Download today it's FREE!>

Express yourself instantly with MSN Messenger! MSN Messenger=20
Express yourself instantly with MSN Messenger! Download today it's FREE!
10/23/2008 11:45:08 PM 133541 articles. 0 followers. Post Follow

0 Replies

Similar Articles

[PageSpeed] 14
Similar Artilces:

Re: ot
A very good all purpose server product of this nature that I have extensi= ve=20 experience with is FirstClass ( * To join/leave the list, search archives, change list settings, * * etc., please visit * ...

Re: Out of Area Calls #17 (Robert Bonomi) writes: > pointed out that the law made *no* such provision for delay on an > -internally-maintained- list, and required that they update my > _customer_record_ with a note that 'customer has ordered us -never- to > make marketing calls to him', and the date/time. Three days later, > somebody "didn't read" the notes, and called me. I promptly demanded > a supervisor, had them read the account 'notes', and asked if they > wanted to pay the statutory $500 minimum, or if I needed to go to &g...

test #2 #6
ignore ...

US-TX-Port Neches: FIELD SUPPORT, Windows 2000/XP,Norton Ghost; 2-3 M (45352332404)
US-TX-Port Neches: FIELD SUPPORT, Windows 2000/XP,Norton Ghost; 2-3 M (45352332404) =================================================================================== Position: FIELD SUPPORT Reference: ZYD00041 Location: Port Neches TX Duration: 2-3M Skills: Detail-oriented - will follow a written procedure at all times. Windows 2000 or Windows XP migration experience. Windows XP desktop support experience. Ability to troubleshoot hardware and network issues. Experience with Norton Ghost ...

Re: Creating a new variable? #4
Hi, Yes, you can do it easily. This is one solution: Proc sort data=3DTab1; by Id; run; Data Tab1; set Tab1; by Id; retain INC 0; if first.Id then INC=3D0; Inc=3Dinc+1; run; Le 16:01 7/02/2006, vous avez =E9crit: >Hi > >I have alarge data set in which I have a unique ID, for each id there are= 49 >observations, now I want to create a new variable which would be equal to 1 >for first observation and increment in steps of 1 to 49 and then reset to 1 >and continue. is there a way of doing this? > >thanks > >Vj Alexandre BAUDRY Unit=E9...

RE: Question to Kerry Main
> -----Original Message----- > From: JF Mezei [] > Sent: June 30, 2007 8:31 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Question to Kerry Main >=20 > Mr Main, your arguments on c.o.v. are often rebutted as part of normal > debating process. (And you rebutt other arguments with your own > responses). >=20 > In real life, do you also find customers and potential customers > having > similar questions ? Or do you conclude that we, in c.o.v. are a > sepcial > bunch living in a totally different universe without a clue...

Re: Really Frustrating -9628 Errors #2
On Thu, 2005-11-17 at 14:23 -0500, Adam Tauno Williams wrote: > We've written several Informix stored procedures that return MULTISET as > the type; the application than does a SELECT a,b,c FROM > TABLE({procedurename}({parameterlist...})). And these WORK!!! But only > for awhile, the next day, or after a couple of days suddenly they stop > working - and they haven't been changed. They simply start to bomb out > with a -9628 (type not found) error for any SQL statement like the > above. > We are using "IBM Informix Dynamic Server Version 10.00...

Re: String question #6
As Howard example shows the last observation passed incorrectly, Is there a reason why the third observation(123 12) did not pass incorrectly. Is that a length issue or something different? On Thu, 1 May 2008 09:50:19 -0400, Howard Schreier <hs AT dc-sug DOT org> <schreier.junk.mail@GMAIL.COM> wrote: >On Thu, 1 May 2008 08:16:37 -0400, Richard A. DeVenezia ><rdevenezia@WILDBLUE.NET> wrote: > >>Dominic Mitchell wrote: >>> input var1 var2 $6.; >>> cards; >>> 99 9982 >>> 998 998 >>> ; >> >>The INPUT statem...

FS: childrens EDUCATIONAL SoftWare #2
We give you MORE than others with FAST efficient despatch and FREE postal insurance. With OUR Full Money back guarantee for faulty software including the cost of returning the item. Visit our store to see our feedback, we have thousands of satisfied customers Find this and more at our fully searchable EBAY shop: For Children's Games, Educational, Language & Recreational Software ...

Re: More SQL simplification
Roger: In your version of the query, SAS 'remerges' summarized values back to the original dataset. It actually works the same as the query that joins a view to the original dataset. The simpler form of the query does not in general across different flavors of SQL. Sig -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Roger Lustig Sent: Tuesday, September 14, 2004 7:12 PM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: More SQL simplification Pat: You've answered your own question! No reason you can't do a GROUP BY while selecting colum...

difference between 2 sounds
Hi How can I show the difference beetwen 2 signals, for example 2 sounds (sinus) thx joe "joe " <> wrote in message <ha08k6$aco$>... > Hi > How can I show the difference beetwen 2 signals, for example 2 sounds (sinus) > > thx > joe Hi Joe, can you be more specific so that someone can help you? Do you mean you want to show differences between to signals (sounds) in terms of the way energy is distributed as a function of frequency, or differences in the time behavior. These are just two possibilities. wayne "...

Re: question about data values #11
Howdy Toby - The V9 compress function has new useful modifiers to handle all numbers, all letters, typewriter characters etc. In this case D is for digits 0-9 and K is for Keep (rather than compress). 10 data _null_ ; 11 x = 'A1234' ; 12 y = input(compress(x,'','DK'),5.0); 13 put x= y= ; 14 run ; x=A1234 y=1234 regards as always - Paul Choate DDS Data Extraction (916) 654-2160 -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of toby dunn Sent: Monday, June 27, 2005 4:41 PM To: SAS-L@...

This is an autoresponder. I'll never see your message. ...

Re: How to obtain FrameTicks List from an existing Plot?
t = Range[5]; p = ListPlot[t, Frame -> True]; FrameTicks /. AbsoluteOptions[p] Bob Hanlon ---- Nasser Abbasi <> wrote: ============= I have a need to obtain the FrameTicks list from an existing plot. The reason I want to do this, is because I need to slightly modify the ticks, and then redraw the plot with the new Ticks. This would be much easier than having to generate the Ticks myself in the first place. This is an example: t = Range[5]; p = ListPlot[t, Frame -> True] But when I look at FullForm[p], I see no list would would look like the...

Re: [HP3000-L] Save and Re-Use a File Equation
Jay writes: > What I'd like to do is to save the file equation with the parameters the > user entered, and then re-use that file equation after the checks are > printed. Like this: > > 1. Enter printer name, etc for reports > 2. Submit job, which does: > a. file report;dev=<parameters from 1 above> > b. run program A > c. file print=checks;dev=lp;forms=mount checks > d. run program B to print checks > e. file report;dev=<parameters from 1 above> > f. run program C to create report to print on printer ...

Re: array label question #9
Here's another approach. conceptually similar to the previous posts..... data _null_; %let colmns = ; do yr = 2007,2008; do qtr = 1 to 4; do channel_cd = 1 to 3; do Incentive_Type = 1,2; call symput ('colmns', symget ('colmns') || ' ' || '_' || catx('_',yr,qtr,channel_Cd,Incentive_Type)); end; end; end; end; run; data transpose (drop = yr qtr channel_cd incentive_type contacts) ; format &colmns best32.; array all [2007:2008, 1:4, 1:3, 1:2] _all_; do until(last.indiv_id) ; set have; by indiv_id; all{yr,qtr,ch...

Re: Question re Canon iP1500 #2
Been busy for a few days, so didn't get a chance to get back here. But need to ask about your message below. "zakezuke" <> wrote in message > > Do y ou mean I could fix the problem of the waste ink absorber myself? I > > thought you can't see it by just looking into the printer > > The waste counter you can reset with the service tool QY9-0066 as Tony > explained. You can get it here >

Re: own package doesnt work correctly?
Martin, Symbols like x and I are set up in the package`Private` context once you put the definition inside the package. So they won't match Global`x or Global`I in a notebook using the package. You could either add the Global` context to these symbols in the package, or better yet pass them as parameters to the rule. partintrulx[x_,I_]:= ..... David Park From: Martin [] Hello, I started to - try to - use packages written on my own, using the Notebook editor, marking the cells as initializa...

Re: Now That They're Mounted ... #9
Correct me if I'm wrong, but MPE does not provide automatic failover in the event a controller dies. Yes, you can connect 2 controllers, but one won't take over for the other in the event of failure. That's an important distinction to recognize. John Lee Vaske Computer Solutions At 03:35 PM 8/18/04 -0400, Gilles Schipper wrote: >At 03:20 PM 2004-08-18, you wrote: >>O.K. two brain farts. Can't have more than 7 logical drives. > > >The mod20 supports up to 8 luns or logical drives. > > >>And you can't split the SCSI, (and u...

Re: Apple's iPod Raises Bar For Digital Music Players
Hey- Check out this great site that is giving away totally FREE iPods! I've joined and I think you should as well. It's a completely legitimate offer, and this company has already given away $4 million in FREE stuff! All you have to do is join, complete an online offer, and refer friends to do the same. That's it! Here is my referral link. To help me get my iPod, click this exact link to join, or copy and paste it into a browser: Monty Solomon wrote: > Apple's iPod Raises Bar For Digital Music Players > By W...

Re: Call System in UNIX #7 1547219
Dear SAS-L-ers, Michael D. Ross posted the following interesting question: > Hi All, > I'm having trouble with the following code in Unix. It > creates the "dump.txt" file in the specified directory; > however, it has 0 records in it. Any suggestions? > > > > %let progs = /temp/mydir; > > data _null_; > command= " ll &progs *.sas > &progs/dump.txt"; > call system (command); > run; > Michael, I see it! It appears that you are falling prey to a subtle problem in your DATA _NULL_ step. Your problem is with the spac...

Nesting multiple show/hide divs
Forgive me, I'm still pretty newb with javascript. I was asked to start updating a webpage for an organization I belong to, and I'd like to make it a bit easier to read. Before I get too far, the current page is located As you can see, it is very long, and pretty confusing. What I'd like to do is for each name level, use a show/hide div for the one under it. For example, all of the headers would always show. But you could click on that, and the level 2 names would show up. Click on a level 2 name, and level 3 names (but only...

OT: Music on Halo 2
The games good but the music on the menu screen is amazing. It's been ages since I noticed the music on anygame but Halo 2's music is fantastic. Anyone else notice this? "Cheddar" <> wrote in message news:4198e0fa$0$36693$ > The games good but the music on the menu screen is amazing. It's been ages > since I noticed the music on anygame but Halo 2's music is fantastic. > > Anyone else notice this? > > Interesting choice of group to ask :P On Mon, 15 Nov 2004 17:39:12 -0000, &quo...

Re: New to SAS #2
Here is a very good paper that covers "everything" about learning more about SAS. It lists many resources of varying types, including SAS-L: -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Sent: January 14, 2008 3:11 PM To: SAS-L@LISTSERV.UGA.EDU Subject: New to SAS I'm new to SAS and Data Extraction generally, does anyone have any hints, tips or advice that might be useful? I'm interested in: related websites, programming languages, programmes, b...

Re: Thanks Harry Re: SAS Data Step Qn.
Use Eric's solution, it's cleaner. -----Original Message----- From: []On Behalf Of Jeff Morison Sent: Tuesday, August 09, 2005 3:03 PM To: Droogendyk, Harry; SAS-L@LISTSERV.UGA.EDU Subject: Thanks Harry Re: SAS Data Step Qn. This worked out perfectly Harry, thanks very much. --- "Droogendyk, Harry" <harry.droogendyk@RBC.COM> wrote: > data a; > input HRN : CDATE : date7. STAGE : > start : stop; > format cdate date9.; > cards; > 609 21JUL03 2 307 ...