Gov2001 April 2011

gov2001@lists.gking.harvard.edu

26 participants
44 discussions

Bivariate probit regression with survey weights

by Alexander Hertel-Fernandez

Hey 2001istas, I'm trying to fit a bivariate probit for survey data. It appears that the bprobit model in Zelig doesn't support survey weights. Does anyone know of an alternative or workaround? Thanks! --Alex -- Alexander Hertel-Fernandez ahertel(a)fas.harvard.edu alexander.hertel(a)gmail.com 765.430.2063 _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l -- gov2001 mailing list served by Harvard-MIT Data Center List Address: gov2001(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=gov2001

13 years

Marathon

by Gary King

Is anyone in class running in the marathon on Monday? we usually have a few. Gary King http://gking.harvard.edu 617-500-7570 (from my phone) _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

web sites

by Gary King

Hi Gang, Its time for you all to have a web site. If you don't, or even if you do, I suggest you have a look at OpenScholar, which is a project we created at IQSS to make this easy. OpenScholar includes a broad (and growing) array of useful and innovative features to choose from, and allows a look, feel, design, and URL of your choice (including the ability to take your site with you when you graduate). We're rolling it out for faculty now, and increasing numbers of faculty and students across the university are adopting it. To create your own site, or to see some of the others already included, go to http://scholar.harvard.edu. (Excuse the cross-post on the gov grad student list..) Gary -- *Gary King* - Albert J. Weatherhead III University Professor - Director, IQSS - Harvard University GKing.Harvard.edu <http://gking.harvard.edu/> - King(a)Harvard.edu - @kinggary<http://twitter.com/kinggary>- 617-500-7570 - Asst 495-9271 - Fax 812-8581 _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

Zero Inflated Models

by Colin Sullivan

Hello 2001-ders, Is "zero-inflated" a general term for mixed models where one distribution is zero? Or does it only apply to count models? For instance, is there a "zero-inflated logit" model for binary outcomes? Thanks, Colin _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

Video Annotation

by Brandon Stewart

Gov2001, As some of you may know the video annotation tool we use for class was developed specifically for our class and is going to be made available to the university at large. The technology office would like to conduct some "simple (no pressure) short interviews" with students. These video testimonials will help them promote their product, get funding from the university etc. If you would be willing to do one of these interviews after class one day, shoot me an email and let me know. Thanks! Brandon _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

First differences in mlogit and ologit using zelig

by Rachel Crouch

Hi class, I'm having some trouble interpreting mlogit and ologit models I am working on for the replication paper. I am able to get the models to run, but then when I try to figure out the first differences, I get an error: >x.noflood <- setx(z.out.mnl.1, FLOOD.2d.v1 = 0) Error in "contrasts<-"(`*tmp*`, value = "contr.treatment") : contrasts can be applied only to factors with 2 or more levels This is odd, because the factor does have more than 2 levels. I think this has been a problem for others in zelig in the past, but none of the dialogue I have found online about it helps. Setting the dependent variable to a factor before entering it into the model, as suggested by some people on help forums doesn't help. Have any of you run across this issue before? Thanks! Rachel _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

matchit error

by Slawa Rokicki

Hey, I am getting this error when I run matchit using cem (for my rep paper): Error in matchit2cem(c(0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, : No control units were matched I am not sure what this means. For my treatment variable I do have both treated and nontreated. Any ideas? Thanks! Slawa _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

Beamer

by Brandon Stewart

Hi Gov2001, I've gotten a couple of questions about how Molly, Gary and I make slides using LaTeX. Since most of you will be teaching at some point in the future, I thought I'd share a few resources to get you started. The short answer is we use the package Beamer. It uses almost exactly the same LaTeX commands with a couple of different flourishes. You will be able to pick it up in no time. Some Resources for You: Simple Bare-Bones Beamer: http://www.math.umbc.edu/~rouben/beamer/quickstart-Z-H-3.html#node_sec_3 A Great Beamer Tutorial in Beamer: http://www.uncg.edu/cmp/reu/presentations/Charles%20Batts%20-%20Beamer%20Tu… Advanced Beamer Tutorial: http://faq.ktug.or.kr/wiki/uploads/beamer_guide.pdf The Beamer Manual (via Gary's website: http://gking.harvard.edu/files/beameruserguide2.pdf The Official Beamer Page: https://bitbucket.org/rivanvx/beamer/wiki/Home Note like all of LaTeX, the first few times you use it you will say "Why am I not using powerpoint! It's so much easier!" but after you get the hang of it, you will never go back. Enjoy! Brandon _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

Multiple Imputation in STATA

by Yue Hou

Hi everyone, I ran into a multiple imputation problem using STATA and am wondering whether you guys have encountered this problem before. I've generated a bunch of datasets using Amelia and went back to STATA to run regressions. The code looks something like this: miest dataset xtpcse Y X1 X2 X3 X4 X5, nsets(10) Strangely, when I apply the (almost) same "miest" function on another set of imputed datasets, STATA returned an error saying "must specify panelvar and timevar; use xtset",which didn't happen when I use the other datasets. And when I use help function to search for syntax for "miest", the pop out says "a file called miest was not found". But when I apply the same code on another set of datasets there's no such problem at all. Do you have any idea what's wrong with it? Many thanks, Yue _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l -- gov2001 mailing list served by Harvard-MIT Data Center List Address: gov2001(a)lists.gking.harvard.edu Subscribe/Unsubscribe: http://lists.gking.harvard.edu/?info=gov2001

13 years

Incompatible arguments in R

by Rachel Crouch

Hi team, Maybe someone has run across this? When I try to draw from mvrnorm, R gives me an "incompatible arguments" error. I thought I would send it out to the list before spinning my wheels any longer. The code is below (the first half or so is just for context). Let me know if you have any thoughts. Thanks! Rachel #Logit model: z.out.1 <- zelig(REVERSAL ~ TREND.2d + TREND.9th + TRANS.2d.v1 + CHANGEt.2d.v1 + FLOOD.2d.v1 + CHANGEf.2d.v1 + TRANS.9th.v1 + CHANGEt.9th.v1 + FLOOD.9th.v1 + CHANGEf.9th.v1 + factor(JUDGQY) + factor(CIRCUIT) + GOP + DCTGOP + VACANCY, model = "logit", data = caselevel.data.logit) #Robust standard errors: temp.lrm <-lrm(z.out.1, x=TRUE, y=TRUE, data=caselevel.data.logit) g1 <- robcov(temp.lrm, caselevel.data.logit$CIRCUIT) #Simulating: se1 <- sqrt(diag(g1$var)) coef1 <- z.out.1$coef se1 <- as.matrix(se1) coef1 <- as.matrix(coef1) set.seed(02138) betas <- mvrnorm(n = 5000, mu = coef1, Sigma = se1) _______________________________________________ gov2001-l mailing list gov2001-l(a)lists.fas.harvard.edu http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l

13 years

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

Gov2001 April 2011