LEGACY CONTENT. If you are looking for Voteview.com, PLEASE CLICK HERE

This site is an archived version of Voteview.com archived from University of Georgia on May 23, 2017. This point-in-time capture includes all files publicly linked on Voteview.com at that time. We provide access to this content as a service to ensure that past users of Voteview.com have access to historical files. This content will remain online until at least January 1st, 2018. UCLA provides no warranty or guarantee of access to these files.

POLI 272 BAYESIAN METHODS
Seventh Assignment
Due 19 November 2009

All homeworks must be neatly typed with Microsoft Word or its equivalent with all R and WINBUGS code shown in an Appendix.

  1. Use the Examples that I have posted of Ernesto Calvo's Interface with WINBUGS to rerun re-run the three Houses you analyzed in Homework 4, Homework 5, and Homework 6. Read the STATA files as shown in several of the Calvo examples and run both Probit and Logit models.

  2. Use the regression example of output From Running Calvo's R2WINBUGS Code on Bush's 2000 Vote By Congressional District (PDF) to run a similar model using this STATA dataset of the 1996 U.S. Presidential election by Congressional district:

    pelect96.dta -- STATA data file for 1996 Presidential Election Results by Congressional Districtions. Here are the variables: 
    . d

Contains data from C:\Inetpub\ftproot\wf1\pelect96.dta
  obs:           407                          
 vars:            10                          4 Oct 2000 11:22
 size:         9,361 (99.1% of memory free)
-------------------------------------------------------------------------------
              storage  display     value
variable name   type   format      label      variable label
-------------------------------------------------------------------------------
clint96         byte   %8.0g                  clinton vote
dole96          byte   %8.0g                  dole vote
perot96         byte   %8.0g                  perot vote
income          float  %9.0g                  distr. median income
nom1            float  %9.0g                  nominate 1st dim.
nom2            float  %9.0g                  nominate 2nd dim.
party           byte   %8.0g                  political party 0=Dem, 1=Rep
south           byte   %8.0g                  region 0=north, 1=south
afram           byte   %8.0g                  percent black
hisp            byte   %8.0g                  percent hispanic
-------------------------------------------------------------------------------
Sorted by:  

. summ

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
     clint96 |       407    50.41769    12.74948         26         94
      dole96 |       407    39.73464    11.77471          4         68
     perot96 |       407    8.289926    2.876689          1         16
      income |       407    30.83125    8.369561     15.052     57.219
        nom1 |       407    .0459017    .3525151      -.625        .71
-------------+--------------------------------------------------------
        nom2 |       407   -.1009656    .4474046      -1.73       1.12
       party |       407    .5380835    .4991611          0          1
       south |       407    .2678133    .4433647          0          1
       afram |       407    10.96314    15.61091          0         74
        hisp |       407    8.722359    14.20137          0         83
    1. Run the regression of the Clinton vote -- clint96 -- on income, party, south, afram, hisp, nom1, nom2. Compare the results to STATA.

    2. Run the regression of the Dole vote -- dole96 -- on income, party, south, afram, hisp, nom1, nom2. Compare the results to STATA.

    3. Run the regression of the Perot vote -- perot96 -- on income, party, south, afram, hisp, nom1, nom2. Compare the results to STATA.

    4. Create a dichotomous variable in STATA that is "1" if Clinton's vote is greater than or equal to 50%, and "0" otherwise. Run the probit model using the same specification as above using the Calvo model.

    5. Run the Logit model for Clinton's vote using the Calvo model.

    6. Calculate the Pearson Correlation Coefficient between the coefficients (excluding the intercept term!!) between the Probit and Logit coefficients for Clinton's vote.

  3. Repeat 2(d)-2(f) for the Dole vote.