---

1. In this problem we are going to try Lewis and Poole's Parametric Bootstrap with WNOMINATE.
   
   Download the R program:
   
   
   wnominate_bootstrap.r -- Program to run the Parametric Bootstrap with WNOMINATE.
   
   sen90kh.ord -- 90^th (1967-68) Senate Roll Call Data
   
   
   1. Run wnominate_bootstrap.r and turn in the plot that it makes.
      
      
   2. Report result$fits.
      
      
   3. Turn in plot(result).
      
      
   4. Change the number of trials to 501 in:
      
      result <- wnominate(hr, dims=2, trials=51, polarity=c(7,2))
      
      repeat (a), (b), and (c).
      
      
2. Continuing with Lewis and Poole's Parametric Bootstrap with WNOMINATE we are going to scale the 90^th (1967-68) House and compare it to the 90^th Senate.
   
   Download the R program:
   
   
   wnominate_bootstrap_h90.r -- Program to run the Parametric Bootstrap with WNOMINATE.
   
   hou90kh.ord -- 90^th (1967-68) House Roll Call Data
   
   
   1. Run wnominate_bootstrap_h90.r and turn in the plot that it makes. Note that this is a two panel plot.
      
      
   2. Report result$fits.
      
      
   3. Turn in plot(result).
      
      
   4. Change the number of trials to 501 in:
      
      result <- wnominate(hr, dims=2, trials=51, polarity=c(2,15))
      
      repeat (a), (b), and (c).
      
      
3. In this problem we are going to compare Simon Jackman's Ideal program for analyzing roll call data with WNOMINATE. Ideal is part of the pscl package that has many nice routines (see this pdf describing the package, 102 pages ).
   
   Download the R program:
   
   
   ideal_90_senate_4.r -- program to analyze the 90^th U.S. Senate roll call votes with Ideal and with WNOMINATE.
   sen90kh.ord -- 90^th U.S. Senate roll call data
   
   
   1. Run ideal_90_senate_4.r and turn in the two plots that it makes. Note that one is a two panel plot and another is a single panel plot.
      
      
   2. Report summary(result_id) (You do not need to report the ideal point summary; just the overall summary with the ideal point means).
      
      
   3. Report summary(resultw), resultw$fits, and resultw$weights.
      
      
   4. Report summary(olsideal2). Note that the command:
      
      olsideal2 <-lm(distij ~ distideal1+distideal2 - 1)
      
      is simple OLS with the intercept suppressed!! Report
      
      cor(distij,distfix)^2
      
      why is this number different than that reported in summary(olsideal2)?
      
      
4. Modify ideal_90_senate_4.r to analyze hou90kh.ord, the 90^th (1967-68) House Roll Call Data. Note that you will need to change the labeling on all the plots! Call the program ideal_90_house_4.r. Note that you will have to fix this command with three appropriately chosen members of the 90^th House:
   
   result_id <- postProcess(result,constraints=list(MONDALE=c(-0.822,-0.043),THURMOND=c(0.994,-0.109),PERCY=c(-0.151,-0.988)))
   
   Note that in wnominate_bootstrap_h90.r you write the NOMINATE coordinates to disk with this command:
   
   write.fwf(x=format(as.data.frame(result$legislators),digits=5,width=10,
   scientific=FALSE),"c:/uga_course_homework_14/wnomtest_x_h90.txt")
   
   Find two legislators who are near -1,0 and +1,0 to anchor the first dimension and find a third legislator who is near 0.0 on the first dimension but has a large 2nd dimension coordinate (either positive or negative). Note that the three names need to be unique because of a bug in Ideal.
   
   
   1. Repeat 3(a), 3(b), 3(c), and 3(d).
      
      

---