The data below includes the individuals (who are identified by ID numbers i = 1,..., 20)...

Transcribed Image Text:

The data below includes the individuals (who are identified by ID numbers i = 1,..., 20) and their monthly wages (in thousands), gender, education levels (years of education), experience (years of work), and the regions where they live. ID 1 2 Wage 17.18 32.73 18.81 33.27 35.50 32.24 7 19.69 8 17.41 9 17.69 10 25.90 11 24.54 12 21.27 13 28.66 14 24.23 15 18.40 3 4 5 6 16 30.48 17 31.41 18 19 20 38.67 33 16.12 Education 8 16 16 12 16 16 16 12 8 16 12 16 12 8 16 16 16 Table 1: Observations 16 12 16 Experience 1 5 11 6 18 15 4 15 5 19 5 U 11 8 4 16 17 12 13 0 1 Gender Female Male Female Male Male Male Female Female Female Female Male Female Male Male Female Male Male Male Male Female Region Aegean Aegean East Anatolia Black Sea Aegean Aegean Black Sea East Anatolia Black Sea Marmara Black Sea Black Sea East Anatolia East Anatolia Aegean East Anatolia Aegean Marmara Aegean Aegean (a) Interpret the data. Do you think eyeballing helps you to make an inference that regional and gender differences are projected in wages? Are there any outliers? (15 Points) (b) Your manager asks you to study the impact of being a college graduate (16 years) and/or high school graduate (12 Years) on earnings relative to having a secondary school diploma. What would be your strategy to estimate this model? Write down the regression model. Construct the dummy variables if it is necessary. (15 Points) (c) Create dummy variables to measure the regional effects and write the first five observations of those dummy variables. How many dummy variables you should create to avoid dummy variable trap? (20 Points) (d) Do you think R2 will be higher if we include more independent variable to the wage model? (15 Points) (e) You are given below the correlation matrix for variables Education, Experience, and Female. Suppose that we are interested in finding the impact of education on earnings. Based on the correlation matrix, should we care about omitted variable bias? If we should, what would be your next action? (15 Points) Education Experience Female Table 2: Correlation Matrix Education Experience 1 0.27 0.27 1 -0.01 -0.40 Female -0.01 -0.40 1 The data below includes the individuals (who are identified by ID numbers i = 1,..., 20) and their monthly wages (in thousands), gender, education levels (years of education), experience (years of work), and the regions where they live. ID 1 2 Wage 17.18 32.73 18.81 33.27 35.50 32.24 7 19.69 8 17.41 9 17.69 10 25.90 11 24.54 12 21.27 13 28.66 14 24.23 15 18.40 3 4 5 6 16 30.48 17 31.41 18 19 20 38.67 33 16.12 Education 8 16 16 12 16 16 16 12 8 16 12 16 12 8 16 16 16 Table 1: Observations 16 12 16 Experience 1 5 11 6 18 15 4 15 5 19 5 U 11 8 4 16 17 12 13 0 1 Gender Female Male Female Male Male Male Female Female Female Female Male Female Male Male Female Male Male Male Male Female Region Aegean Aegean East Anatolia Black Sea Aegean Aegean Black Sea East Anatolia Black Sea Marmara Black Sea Black Sea East Anatolia East Anatolia Aegean East Anatolia Aegean Marmara Aegean Aegean (a) Interpret the data. Do you think eyeballing helps you to make an inference that regional and gender differences are projected in wages? Are there any outliers? (15 Points) (b) Your manager asks you to study the impact of being a college graduate (16 years) and/or high school graduate (12 Years) on earnings relative to having a secondary school diploma. What would be your strategy to estimate this model? Write down the regression model. Construct the dummy variables if it is necessary. (15 Points) (c) Create dummy variables to measure the regional effects and write the first five observations of those dummy variables. How many dummy variables you should create to avoid dummy variable trap? (20 Points) (d) Do you think R2 will be higher if we include more independent variable to the wage model? (15 Points) (e) You are given below the correlation matrix for variables Education, Experience, and Female. Suppose that we are interested in finding the impact of education on earnings. Based on the correlation matrix, should we care about omitted variable bias? If we should, what would be your next action? (15 Points) Education Experience Female Table 2: Correlation Matrix Education Experience 1 0.27 0.27 1 -0.01 -0.40 Female -0.01 -0.40 1

Answer rating: 100% (QA)

a Eyeballing the data we can see that there are some differences in wages by region and gender The highest wages are earned by individuals in the Marmara and East Anatolia regions while the lowest wag... View the full answer