Part I: Multiple Choice (Worth 25 points). Select the option that best answers t
ID: 1106173 • Letter: P
Question
Part I: Multiple Choice (Worth 25 points). Select the option that best answers the question. Consider the following theoretical models: Model 1.Y = .xf1xf. What models can be linearized and estimated via OLS? 1. Model 2: Y = ° + AW, + .x2 + 2X3 Model 3-Y-A+ a. Only model 1 b. Only model 2 c. Only model 3 d. Both model 1 and model 3 e. Both model 2 and model 3 2. A researcher wants to include a categorical variable for marital status: Married, Single; and a categorical variable for ethnicity: White, Black, Hispanic. model? How many intercept dummy variables should be added in the b. 3 C. 4 d. 5 It depends on what the reference category is e. 3. When we include a slope dummy variable with two categories, we are a. b. c. d. e. changing the intercept of the estimated equation changing the slope of the estimated equation changing the reference category of the estimated equation adding one dummy variable to the estimated equation making the model nonlinear in the parameters Consider the following model:Y-Po + Ax, + Axl + Axate. The marginal effect of X, on Y is 4. a. B e. none of the above If we swap the reference category in a model with an intercept dummy variable, say replacing the M with the FEMALE dummy, 5. a. The R2 of the equation is unaffected b. The R of the equation changes in value c. The t-statistic of the dummy variable changes signs d. Both a. and c. are right e. Both b. and c. are rightExplanation / Answer
1) a is correct
Model 2 is already a linear model. Model 3 is a reciprocal modern cannot be linearlised. Model 1 can be linearlised by taking logs on both sides.
2) b is correct.
To avoid dummy variable trap we include n-1 dummy variable for categorical variables where n is number of categories of each variable. For marital status we include 1 variable and for ethinicity we include 2 variables. In total 3 variables are included.
3) d is correct
As discussed in question 2 we include n-1 variables. If we have two categories we include one variable.
4) d is correct.
the marginal effect of X2 on Y is the differentiation of Y with respect to X2 which is equal to beta 3.