DA 292 #3

• 
  

  False


  In a regression model where the dependent variable is percentage of voter turnout, and the independent variables are:

  
  

  State party affiliation (1 = Republican State, 0 = Democratic),

  Percentage of population who are registered Democrats between ages 18-40.

  Percentage of population who are registered Republicans between ages 45-70.

  Total number of registered voters by state

  
  

  You would not expect multicollinearity.
• False


  Increasing the sample size will remove/fix multicollinearity problems in the data.
• True


  Including dummy variables to a model will likely increase R-square
• True


  Increasing the number of continuous quantitative variables in a model may decrease SSE.
• True


  For the following model Y = f(X1,X2,X3), the calculated Variance Inflation Factor(VIF) was 8.2. You should expect an R-square > 0.7.
• Decrease on average by 50


  For multiple regression model predicted y = 50 + 25x subscript 1 - 10x subscript 2 + 8x subscript 3, if x subscript 2 were to increase by 5, holding x subscript 1 and x subscript 3 constant, the value of y would: