Zelig February 2009

zelig@lists.gking.harvard.edu

10 participants
17 discussions

by Kosuke Imai

hmmm... I don't use JGR... I'm ccing to the list in case someone knows how to fix this problem. Kosuke -- Department of Politics Princeton University http://imai.princeton.edu On Tue, 17 Feb 2009, Tracy Lightcap wrote: > Kosuke (if I may): > > Here's another minor problem. I run JGR, mainly for the editing help. Zelig > loads correctly, but you get the blue beachball when you try to run demo > (ls). When you try to stop R, the demos, perversely, begin to show up until R > decides it has to truncate things. > > Everything works like a charm on the usual OSX R GUI. Probably a JAVA > problem. > > System: 2.33 Intel Powermac, 2GB RAM, OSX 10.5.6/ JGR 1.6.2. > > Tracy > > On Feb 17, 2009, at 1:09 PM, Kosuke Imai wrote: > >> I have a mac too, but I wonder if you did not install it correctly. You >> can install it via: >> >> install.packages("Zelig", repos = "http://cran.r-project.org") >> >> at the R prompt. Then, try >> >> library(Zelig) >> demo(ls) >> >> If that command runs, then you are good to go. >> >> Kosuke >> >> -- >> Department of Politics >> Princeton University >> http://imai.princeton.edu >> >> On Tue, 17 Feb 2009, Hillard Pouncy wrote: >> >>> Kosuke: >>> >>> thanks. >>> >>> I am using R version 2.8.1 and I did a check for updates. >>> >>> I think I updated Zelig this morning - version 3.4-3. I installed using >>>> source () and Library (Zelig) commands and for good measure I downloaded >>> version 3.4-3 directly from the website and plugged into my applications >>> folder (I am on a Mac) and did another Library(Zelig). >>> >>> I did an equation using R commands and got a "call" >>> >>>> turnc<-data.frame(x=PREP100,y=PREP100) >>>> fm<-lm(y~x,data=turnc) >>>> summary(fm) >>> >>> Call: >>> lm(formula = y ~ x, data = turnc) >>> >>> Residuals: >>> Min 1Q Median 3Q Max >>> -3.71e-13 -1.79e-16 2.27e-16 5.55e-16 1.01e-14 >>> >>> Coefficients: >>> Estimate Std. Error t value Pr(>|t|) >>> (Intercept) 1.11e-14 3.55e-16 3.12e+01 <2e-16 *** >>> x 1.00e+00 1.58e-17 6.31e+16 <2e-16 *** >>> --- >>> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 >>> >>> Residual standard error: 9.12e-15 on 1679 degrees of freedom >>> Multiple R-squared: 1, Adjusted R-squared: 1 >>> F-statistic: 3.99e+33 on 1 and 1679 DF, p-value: <2e-16 >>> >>> >>> I then tried the same data and variables on Zelig (as reinstalled) and got >>> no 'call.' >>> >>> z<-zelig(y~x,model="ls",data=turnc) >>> Error in zelig.default(y ~ x, model = "ls", data = turnc) : >>> no slot "call" in the result >>> >>> I did a names() check and in data turnc, an x and y variable are there. >>> >>> Am I not installing the updated Zelig properly? >>> >>> HP >>> >>> On Tue, Feb 17, 2009 at 8:30 AM, Kosuke Imai <kimai(a)princeton.edu> wrote: >>> >>>> I think you need to upgrade your Zelig (as well as R) to the most current >>>> version. >>>> >>>> Kosuke >>>> >>>> --------------------------------------------------------- >>>> Kosuke Imai Office: Corwin Hall 041 >>>> Assistant Professor Phone: 609-258-6601 >>>> Department of Politics Fax: 973-556-1929 >>>> Princeton University Email: kimai(a)Princeton.Edu >>>> Princeton, NJ 08544-1012 http://imai.princeton.edu >>>> --------------------------------------------------------- >>>> >>>> >>>> >>>> On Feb 17, 2009, at 2:38 AM, Hillard Pouncy wrote: >>>> >>>> I am a Zelig newcomer. I ran my first regression and got the following >>>>> error message. >>>>> >>>>> Error in zelig.default(PTURN100 ~ PREP100 + blk100, model = "ls", data = >>>>> turnb) : >>>>> no slot "call" in the result >>>>> >>>>> Suggestions please. >>>>> >>>>> best >>>>> >>>> >>>> >>> >> >> - >> Zelig Mailing List, served by Harvard-MIT Data Center >> Send messages: zelig(a)lists.gking.harvard.edu >> [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig >> Zelig program information: http://gking.harvard.edu/zelig/ > > Tracy Lightcap > Professor and Chair > Department of Political Science > LaGrange College > 601 Broad St. > LaGrange GA 30240-2999 > (O) tlightcap(a)lagrange.edu (H) altlamp(a)mindspring.com > 706.880.8226 > www.lagrange.edu/academics/political-science/faculty/tlightcap.aspx > > > - Zelig Mailing List, served by Harvard-MIT Data Center Send messages: zelig(a)lists.gking.harvard.edu [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig Zelig program information: http://gking.harvard.edu/zelig/

15 years, 3 months

Re: [zelig] error message

by Kosuke Imai

I have a mac too, but I wonder if you did not install it correctly. You can install it via: install.packages("Zelig", repos = "http://cran.r-project.org") at the R prompt. Then, try library(Zelig) demo(ls) If that command runs, then you are good to go. Kosuke -- Department of Politics Princeton University http://imai.princeton.edu On Tue, 17 Feb 2009, Hillard Pouncy wrote: > Kosuke: > > thanks. > > I am using R version 2.8.1 and I did a check for updates. > > I think I updated Zelig this morning - version 3.4-3. I installed using >> source () and Library (Zelig) commands and for good measure I downloaded > version 3.4-3 directly from the website and plugged into my applications > folder (I am on a Mac) and did another Library(Zelig). > > I did an equation using R commands and got a "call" > >> turnc<-data.frame(x=PREP100,y=PREP100) >> fm<-lm(y~x,data=turnc) >> summary(fm) > > Call: > lm(formula = y ~ x, data = turnc) > > Residuals: > Min 1Q Median 3Q Max > -3.71e-13 -1.79e-16 2.27e-16 5.55e-16 1.01e-14 > > Coefficients: > Estimate Std. Error t value Pr(>|t|) > (Intercept) 1.11e-14 3.55e-16 3.12e+01 <2e-16 *** > x 1.00e+00 1.58e-17 6.31e+16 <2e-16 *** > --- > Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 > > Residual standard error: 9.12e-15 on 1679 degrees of freedom > Multiple R-squared: 1, Adjusted R-squared: 1 > F-statistic: 3.99e+33 on 1 and 1679 DF, p-value: <2e-16 > > > I then tried the same data and variables on Zelig (as reinstalled) and got > no 'call.' > > z<-zelig(y~x,model="ls",data=turnc) > Error in zelig.default(y ~ x, model = "ls", data = turnc) : > no slot "call" in the result > > I did a names() check and in data turnc, an x and y variable are there. > > Am I not installing the updated Zelig properly? > > HP > > On Tue, Feb 17, 2009 at 8:30 AM, Kosuke Imai <kimai(a)princeton.edu> wrote: > >> I think you need to upgrade your Zelig (as well as R) to the most current >> version. >> >> Kosuke >> >> --------------------------------------------------------- >> Kosuke Imai Office: Corwin Hall 041 >> Assistant Professor Phone: 609-258-6601 >> Department of Politics Fax: 973-556-1929 >> Princeton University Email: kimai(a)Princeton.Edu >> Princeton, NJ 08544-1012 http://imai.princeton.edu >> --------------------------------------------------------- >> >> >> >> On Feb 17, 2009, at 2:38 AM, Hillard Pouncy wrote: >> >> I am a Zelig newcomer. I ran my first regression and got the following >>> error message. >>> >>> Error in zelig.default(PTURN100 ~ PREP100 + blk100, model = "ls", data = >>> turnb) : >>> no slot "call" in the result >>> >>> Suggestions please. >>> >>> best >>> >> >> > - Zelig Mailing List, served by Harvard-MIT Data Center Send messages: zelig(a)lists.gking.harvard.edu [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig Zelig program information: http://gking.harvard.edu/zelig/

15 years, 3 months

error message

by Hillard Pouncy

I am a Zelig newcomer. I ran my first regression and got the following error message. Error in zelig.default(PTURN100 ~ PREP100 + blk100, model = "ls", data = turnb) : no slot "call" in the result Suggestions please. best

15 years, 3 months

Don't understand output from simulation of mixed model

by Prew, Paul

> Hello, just this week I began looking at zelig. I have analyzed a > mixed model of my own data by following the least squares mixed model > vignette. As soon as I got to the step below, I figured I wasn't > performing the analysis correctly. I readily admit I don't understand > what the 'setx' and 'sim' commands are doing. Thought I would run > some data that I understand, then see how zelig presents the findings. > In general, interpreting and explaining mixed model results is a > problem for me --- I'm curious whether Zelig can make the results more > interpretable. Further down, I have included all the steps in the > analysis. > > Any insight is much appreciated. > > Thanks, Paul > > > DB.out <- setx(DB.out1, fn = NULL) > > > str(DB.out) > 'data.frame': 84 obs. of 7 variables: > $ (Intercept) : num 1 1 1 1 1 1 1 1 1 1 ... > $ Formula[T.0.5D+800H]: num 1 1 1 1 1 1 1 1 1 1 ... > $ Formula[T.1X ] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2 X ] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2D+400H] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2D+800H] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.L2] : num 0 0 0 0 0 0 0 0 0 0 ... > > > I apologize if the pasted output hasn't maintained its structure. I > don't know much about submitting to mailing lists. > > Analysis > I'm looking at the effect of a catalyst DB on the response SR = soil > removal. Seven different formulations are modeled, and RWash = random > wash is the random effect, signifying that 3 experimental units are > measured from each Wash. Therefore, the 3 EU's are related within the > same Wash. > > > str(DB) > 'data.frame': 84 obs. of 4 variables: > $ RWash : Factor w/ 28 levels "1","2","3","4",..: 1 1 1 2 2 2 3 3 3 > 4 ... > $ SR : num 48.1 52.9 51 47.9 53.6 ... > $ LogSR : num 1.68 1.72 1.71 1.68 1.73 ... > $ Formula: Factor w/ 7 levels "0.5D+400H",..: 2 2 2 2 2 2 2 2 2 2 ... > > DB.out1 <- zelig( SR~ Formula + tag(1 | RWash), data = DB, model = > "ls.mixed") > > summary(DB.out1) > > > summary(DB.out1) > Linear mixed model fit by REML > Formula: SR ~ Formula + tag(1 | RWash) > Data: DB > AIC BIC logLik deviance REMLdev > 500 522 -241 508 482 > Random effects: > Groups Name Variance Std.Dev. > RWash (Intercept) 22.3 4.72 > Residual 15.6 3.95 > Number of obs: 84, groups: RWash, 28 > > Fixed effects: > Estimate Std. Error t value > (Intercept) 35.86 2.62 13.67 > Formula[T.0.5D+800H] 18.82 3.71 5.07 > Formula[T.1X ] -23.56 3.71 -6.35 > Formula[T.2 X ] -5.14 3.71 -1.39 > Formula[T.2D+400H] 17.49 3.71 4.71 > Formula[T.2D+800H] 26.75 3.71 7.21 > Formula[T.L2] 28.24 3.71 7.61 > > Correlation of Fixed Effects: > (Intr) F[T.0. F[T.1P F[T.XP F[T.2DB+4 F[T.2DB+8 > F[T.0.5DB+8 -0.707 > Fr[T.1X] -0.707 0.500 > Fr[T.2X] -0.707 0.500 0.500 > F[T.2DB+400] -0.707 0.500 0.500 0.500 > F[T.2DB+800] -0.707 0.500 0.500 0.500 0.500 > F[T.L2] -0.707 0.500 0.500 0.500 0.500 0.500 > > > DB.out <- setx(DB.out1, fn = NULL) > > > str(DB.out) > 'data.frame': 84 obs. of 7 variables: > $ (Intercept) : num 1 1 1 1 1 1 1 1 1 1 ... > $ Formula[T.0.5D+800H]: num 1 1 1 1 1 1 1 1 1 1 ... > $ Formula[T.1X ] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2 X ] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2D+400H] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.2D+800H] : num 0 0 0 0 0 0 0 0 0 0 ... > $ Formula[T.L2] : num 0 0 0 0 0 0 0 0 0 0 ... > > > s.out <- sim(DB.out1, x=DB.out) > > > str(s.out) > List of 7 > $ x , x1 > , call , > zelig.call , par > , qi$ev , qi$pr > : num [1:84, 1:7] 1 1 1 1 1 1 1 1 1 1 ... > ..- attr(*, "dimnames")=List of 2 > .. ..$ : chr [1:84] "1" "2" "3" "4" ... > .. ..$ : chr [1:7] "(Intercept)" "Formula[T.0.5D+800H]" > "Formula[T.1X ]" "Formula[T.2 X ]" ... > $ NA: NULL > $ NA: language sim(object = DB.out1, x = DB.out, num = 1000) > $ NA: language zelig(formula = SR ~ Formula + tag(1 | RWash), model = > "ls.mixed", data = DB) > $ NA:List of 3 > ..$ betas : num [1:1000, 1:7] 39.2 38 41.5 35.9 31.4 ... > .. ..- attr(*, "dimnames")=List of 2 > .. .. ..$ : NULL > .. .. ..$ : chr [1:7] "(Intercept)" "Formula[T.0.5D+800H]" > "Formula[T.1X ]" "Formula[T.2 X ]" ... > ..$ gammas:List of 1 > .. ..$ RWash: num [1:1000, 1] 7.491 -0.218 -0.151 -3.986 -2.02 ... > .. .. ..- attr(*, "dimnames")=List of 2 > .. .. .. ..$ : NULL > .. .. .. ..$ : chr "(Intercept)" > ..$ scale : Named num 3.95 > .. ..- attr(*, "names")= chr "sigmaREML" > $ NA:List of 2 > ..$ ev: num [1:1000, 1:84] 51.7 52.3 52.6 55.6 54.4 ... > .. ..- attr(*, "dimnames")=List of 2 > .. .. ..$ : NULL > .. .. ..$ : chr [1:84] "1" "2" "3" "4" ... > ..$ pr: num [1:1000, 1:84] 62.9 49.5 50.9 51 47.2 ... > $ NA:List of 2 > ..$ ev: chr "Expected Values: E(Y|X)" > ..$ pr: chr "Predicted Values: Y|X" > - attr(*, "class")= chr "zelig" > > > plot(s.out) > This plot didn't look like much, no surprise to you. Just a pair of > vertical lines at 0.0 and 1.0 on the horizontal axis, extending from > 50 to 60 on the vertical axis (Expected values) > CONFIDENTIALITY NOTICE: This e-mail communication and any attachments may contain proprietary and privileged information for the use of the designated recipients named above. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.

15 years, 3 months

Re: Zelig related

by Kosuke Imai

Hi, I think that the problem is you have too many variables on the RHS and thus the model parameters are not identified perhaps due to the collinearity. Given the number of observations you have, I would focus on a smaller model with only a few main variables. Kosuke -- Department of Politics Princeton University http://imai.princeton.edu On Mon, 9 Feb 2009, Roopesh Ranjan wrote: > Hello, > > My name is Roopesh. I am a student at UW. > I have been looking at the zelig function in R to do some computation. > > Here is the code I am using: > > z.out=zelig(count ~ species + site + t_period + species*t_period, > model="poisson.gee",id="unique_id",data=flowers07,robust=TRUE,corstr="unstructured") > > When I run this model with less variables it runs, but I have now 100 > variables in the model including interactions on about 1000 observations. > This time the model doesn't run. It runs with 10-20 variables. Do you have > any idea how to go about it ? > > > Any help would be much appreciated, > > > Thanks, > Roopesh > - Zelig Mailing List, served by Harvard-MIT Data Center Send messages: zelig(a)lists.gking.harvard.edu [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig Zelig program information: http://gking.harvard.edu/zelig/

15 years, 3 months

Downgrade to 3.4-0?

by Shige Song

I have Zelig 3.4-1 installed on my computer, but now the version number at the official home page is 3.4-0, something happened to the new version? Shige

15 years, 3 months

Re: Windows binary for v3.4.1

by Kosuke Imai

Unfortunately, I don't use Windows but it appears that you may be able to use "instlib" option in the "install.packages" command. Check the online help file of that command too. Kosuke --------------------------------------------------------- Kosuke Imai Office: Corwin Hall 041 Assistant Professor Phone: 609-258-6601 Department of Politics Fax: 973-556-1929 Princeton University Email: kimai(a)Princeton.Edu Princeton, NJ 08544-1012 http://imai.princeton.edu --------------------------------------------------------- On Feb 1, 2009, at 4:22 PM, David Pain wrote: > Hi > > I´m using Vista, so I´m stymied by User Account Control, because it > defaults to the protected zone. > > I suspect there's an argument you can add that can specify a folder > outside that zone. > > I'm rusty on R - could you be kind and tell me the trick, please? > ----- Original Message ----- From: "Kosuke Imai" <kimai(a)Princeton.EDU> > To: "David Pain" <dsp(a)engl22.fsnet.co.uk> > Cc: <zelig(a)lists.gking.harvard.edu> > Sent: Saturday, January 31, 2009 2:25 PM > Subject: Re: Windows binary for v3.4.1 > > >> >> You can install it via: >> >> install.packages("Zelig", repos = "http://gking.harvard.edu") >> >> Cheers, >> Kosuke >> >> -- >> Department of Politics >> Princeton University >> http://imai.princeton.edu >> >> On Sat, 31 Jan 2009, David Pain wrote: >> >>> Hi Kosuke >>> >>> The CRAN site has the binary for 3.3.1 - the download link on the >>> Zelig site hompage is to the source code, and I can't see any >>> links on that site to the 3.4.1 binary. >>> >>> Could you point me in the right direction, please? >>> >>> Regards >>> >>> David Pain > > - Zelig Mailing List, served by Harvard-MIT Data Center Send messages: zelig(a)lists.gking.harvard.edu [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig Zelig program information: http://gking.harvard.edu/zelig/

15 years, 3 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

Zelig February 2009