Reference no: EM13183554
Q1. Import the MS Excel "RESTAT Olympic Data" set and produce the standard descriptive statistics for the dataset.
Q2. The variable labels simply repeat the variable names. Relabel the variables so it's clear what each represents. Pay particular attention to the dummy variables.
Q3. Calculate medal shares for total medals won, for each country, by year. The medal share for country in a year is the proportion of the
total number of medals awarded in that year's Olympics that is awarded to the country. To calculate the medal shares, first use an egen function to calculate the total number of medals won in each Olympic year. Then you can generate the yearly medal shares for each country. Label the new variable appropriately.
Q4. Use the tabstat command to check that your calculation of the new medal share variable is correct. Tabulate the the total medals won and the medal share by year. Include the the following statistics: the number of observations, the sum, the minimum, the maximum, and the mean.
Q5. For the the new medal share variable, the sum of the country shares of medals won should be one for each year. State ithat this is true for your new share variable. (This is the main check that you have defined the new variable properly.)
Q6. In the 1960 Olympics, how many medals were awarded in total? What was the maximum number of medals won by any country? What was that country's share (proportion) of medals won?
Q7. In the paper, the researchers indicate that GDP per capita is a key explanatory variable for a country's performance in the Olympics.
Create a new variable that shows GDP per capita for all observations, by dividing gdp by population. Name and label this variable appropriately.
Q8. Create a dummy variable for whether or not each country won any medals at each Olympics. Do not include observations for which the
country's medals won is missing. Name and label this variable appropriately.
Q9. To check this calculation, calculate the standard summary statistics for total medals won, by the categories of your new dummy variable. The calculation should confirm that for the zero category of the dummy variable, no medals were won. Does it? What was the average number of medals won, for those countries that won medals?
Q10. Run a regression that estimates the probability that a country won any medals.The dependent variable should be your dummy variable for whether a country won any medals. The independent variables should include your gdp per capita variable and dummy variables for each Olympic year. To get these year dummies,create them using the i. prefix with the year variable.
Q11. What percent of the variation in the dependent variable does the regression explain?
Q12. In the above regression, is the coefficient of the gdp per capita variable statistically significant at the 5% level? Are the coefficients of
any of the year dummy variables significantly different from zero? If so, which ones? Explain all of this in terms of the t-statistics.
Q13. Calculate the mean gdp per capita in 1996 (a standard summarize command will do this). You will use this calculation in the remainder of the quiz.
Q14. What is the estimated probability that a country with this gdp per capita won a medal in 1996? You may calculate this "by hand" using
the stata display command. In any case, show the calculation by using the relevant coefficients of the regression equation and the relevant values of the right hand side variables.
Q15. Now run a similar regression with total medals won as the dependent variable and the same variables on the right hand side as in the
model above.
Q16. In the above regression, is the coefficient of the gdp per capita variable significantly different from zero at the 5% level?
Are the coefficients of any of the year dummy variables statistically significantly different from zero? If so, which ones? Explain all this in
terms of the p-values.
Q17. What is the estimated number of medals that a country with the average gdp per capita estimated above would have won in 1996? You may calculate this "by hand" using the Stata display command. In any case, show the calculation by using the relevant coefficients of the regression equation and the relevant values of the right-hand side variables.
Q18. Now estimate the same regression as just above, but including only the countries that actually won medals in each year.
Q19. In the above regression, is the coefficient of the gdp per capita variable statistically significant at the 5% level?
Are the coefficients of any of the year dummy variables statistically significant? If so, which ones?
Q20. What is the estimated number of medals that a country with the average gdp per capita estimated above won in 1996? You may calculate this "by hand" using the stata display command. In any case, show the calculation by using the relevant coefficients of the regression equation and the relevant values of the right hand side variables.
Q21. Compare the estimated number of medals won in this regression with the number estimated in the regression above (Q17).
Does the answer make sense; why or why not?