I apologize. Everything works Ok. I mistakenly interpreted the message as a warning message. Sincerely, Olena Ageyeva From: eageeva at hotmail.com To: gov2001-l at lists.fas.harvard.edu Date: Tue, 10 Mar 2009 12:56:36 -0400 Subject: Re: [gov2001-l] zelig() Thanks, Brandon. I have tried to make changes in my code as you suggested, but I still can't make it work. Here is my code: load("shuttle.RData") temperature<-as.numeric(data[,3])#separating variables pressure<-as.numeric(data[,4]) incident<-as.numeric(data[,2]) data<-as.data.frame(cbind(incident,temperature, pressure))#making a data frame z.out<-zelig(incident~temperature+pressure,model="logit", data= data) Running this code I get just the following message and no results: How to cite this model in Zelig: Kosuke Imai, Gary King, and Oliva Lau. 2008. "logit: Logistic Regression for Dichotomous Dependent Variables" in Kosuke Imai, Gary King, and Olivia Lau, "Zelig: Everyone's Statistical Software," http://gking.harvard.edu/zelig I did read the documentation about using zelig(), but it's still not clear to me how to use it. Sincerely, Olena Ageyeva Date: Tue, 10 Mar 2009 07:39:15 -0500 From: bstewart at fas.harvard.edu To: gov2001-l at lists.fas.harvard.edu Subject: Re: [gov2001-l] zelig() Olena, data.frame is an object class in R (all objects have classes in R) that is similar to a matrix but more flexible. Your class is "numeric" after you have done as.numeric. To get a dataframe, you need to combine all the different elements into one dataset. See the following code as an example: shuttle <- load("shuttle.RData")datatbc <- as.numeric(data) ##Data to be convertedIncident <- datatbc[24:46] ##Make VariablesTemperature <- datatbc[47:69] Pressure <- datatbc[70:92]data <- as.data.frame(cbind(Incident, Temperature, Pressure)) ##Combine to make data framecheck <- zelig(Incident ~ Temperature + Pressure, model="logit", data=data) I converted the data, into a numeric vector (the object datatbc) and then sliced that object into variables (the next three lines). Then by combining the three variables with cbind() we create a matrix, that gets converted with "as.data.frame" Finally, on the last line, I've added the relevant zelig code. I could have labeled that variables data$Incident, data$Temperature etc. Specifying the "data" field in Zelig (or most any other model in R) simply allows you to refer to objects with their variable name inside the data frame rather than the dataframe, a dollar sign, the variable name. Brandon On Tue, Mar 10, 2009 at 5:40 AM, Olena Ageyeva <eageeva at hotmail.com> wrote: Have anyone tied it yet? What should be set as "data" parameter? I keep getting an error message and have no idea how to fix it. Error in model.frame.default(formula = as.numeric(incident) ~ as.numeric(temperature) + : 'data' must be a data.frame, not a matrix or an array Sincerely, Olena Ageyeva Windows Live?: Keep your life in sync. Check it out. _______________________________________________ gov2001-l mailing list gov2001-l at lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l Access your email online and on the go with Windows Live Hotmail. Sign up today. _________________________________________________________________ Windows Live?: Life without walls. http://windowslive.com/explore?ocid=TXT_TAGLM_WL_allup_1a_explore_032009

There is a way to do this manually on your computer. Go to the R subdirectory on your computer and find a directory called "etc". It might look something like this for Windows: C:\Program Files\R\R-2.8.1\etc\ Find a file called "Rprofile.site". Open it with a text editor (Notepad for example). Add in the following line to automatically load the Zelig package: options(defaultPackages = c(getOption("defaultPackages"), "Zelig")) You can do this with as many packages as you want, for example: options(defaultPackages = c(getOption("defaultPackages"), "foreign", "lattice")) Save the file. Now everytime you open R, it should automatically load those packages. On Tue, Mar 10, 2009 at 1:54 PM, Allen, Abigail <aallen at hbs.edu> wrote: -- Patrick Lam Department of Government and Institute for Quantitative Social Science, Harvard University http://www.people.fas.harvard.edu/~plam