Exercise 1. Find and download the National Accounts data from the Penn World Tables, version 9.1. Drop all observations for the years before 2000.

1. a)  Create a dummy variable which equals 1 for American countries and 0 for all others. The American countries are: Argentina, Bolivia, Brazil, Canada, Chile, Colombia, Ecuador, Guatemala, Honduras, Mexico, Peru, Paraguay, El Salvador, Surinam, Uruguay, Venezuela, USA. Create a variable called “gdp” which is defined as Real GDP in 2011 constant prices divided by the number of persons engaged. Label this variable as “Real GDP per worker”. Provide the mean and standard deviations for these two new variables for the year 2012.
2. b)  For this part, you might find it useful to create a dummy variable to differentiate countries with “Real GDP per worker” of less than 20,000. Let’s call countries with less than 20,000 real gdp per worker poor countries. Using tabulations and only the year 2012, what fraction of countries are American countries? What fraction of countries are poor? What fraction of countries were American and poor? What fraction of American countries were poor?
3. c)  Construct the 90, 95, and 99% confidence intervals for the proportion of poor countries for the year 2012. Explain the steps required to construct a confidence interval.
4. d)  Test whether real GDP per worker in American countries is different from the real GDP per worker in the rest of the world for the year 2012. Explain all the steps required to conduct hypothesis testing.

 

Exercise 2. Use the data health2015.dta.

1. a)  Create a dummy variable “labour” which is equal to 1 for everyone in the labour force, and 0 for those who are not in the labour force. Anyone who is employed, self-employed, unemployed or on parental leave is in the labour force. How many observations are in the labour force, and how many are not? Make sure to exclude missing values. Tabulate the variables labour and sex and show the results. If an individual is female, what is the probability that she is in the labour force?
2. b)  What are the labour force participation rates among men and among women? Is there a significant difference between the two at the 99% significance level?
3. c)  Among those who have positive labour income, is there a statistically significant difference (at the 95% level) in the average labour income between men and women?
4. d)  Among those who have positive labour income, is there a statistically significant difference (at the 95% level) in the variance of labour income between men and women?