Reference no: EM1321613

Q1) During its manufacture, product is subjected to 4 different tests in sequential order. Efficiency expert claims that fourth (and last) test is needless as its results can be forecasted based on first three tests. To test this claim, multiple regression will be utilized to model Test4 score (y), as function of Test1 score (x1), Test 2 score (x2), and Test3 score (x3).

E(y) = β ₀+ β ₁x₁ + β ₂x₂ + β ₃x₃

Global F statistic is utilized to test null hypothesis, H₀: β ₁ = β ₂ = β ₃ = 0. Explain this hypothesis in words.

a) First three test scores are poor predictors of Test4 score.

b) First three test scores are reliable predictors of Test4 score.

c) Model is not statistically useful for predicting Test4 score.

d) Model is statistically useful for predicting Test4 score.

Q2) When using model E(y) = β ₀ + β ₁x for one qualitative independent variable with a 0 - 1 coding convention, β 1 signifies difference between mean responses for level assigned value 1 and base level.

True

False

Q3) When modelling E(y) with single qualitative independent variable, number of 0 - 1 dummy variables in model is equal to number of levels of qualitative variable.

True

False

Q4) Elections officer wants to model voter turnout (y) in a precinct as function of type of election, national or state.

Write down a model for mean voter turnout, E(y), as function of type of election.

a) E(y) = β 0 + β 1x, where x = 1 if national, 0 if state

b) E(y) = β 0 + β 1x, where x = voter turnout

c) E(y) = β 0 + β 1x1 + β 2x2, where x1 = 1 if national, 0 if not and x2 = 1 if state, 0 if not

d) E(y) = β 0 + β 1x + β 2x2, where x = voter turnout