Econometrics Questions - For all questions you should provide screen print outs of STATA in an appendix where necessary to support your workings.

Question 1 - The file Productivity.xlxs contains cross-sectional information on US coal mines. It in- cludes the following variables:?id = mine identifier?tonshr = number of tons of coal produced per hour (productivity) capital = total value of the machinery stock?dayslost = number of days the mine had to be shutdown over the year to make repairs following accidents?price = the price per ton of the mine's coal vertint = a dummy variable indicating whether the mine is vertically integrated (equal to one if the mine is owned by an electricity power plant, 0 otherwise)?union = a dummy variable indicating whether the mine's workers are unionized (equal to one if the mine is unionized, 0 otherwise)

Over the past two decades the industrial economics literature has sought to explain difference in firms' productivity levels. Ultimately, productivity determines a nation's living standards as it is the key driver of economic growth. To investigate the determinants of productivity in the coal mining sector, estimate the following equation

tonshri = β₀ + β₁capitali + β₂dayslosti + β₃pricei + β₄vertinti + β₅unioni + ε_i          (1)

Using STATA to estimate the OLS estimators of β₀, β₁, β₂, β₃, β₄, and β₅ -

1. Comment on the sign and significance of the coefficient estimates. Provide economic intuition on the relationships you uncover.

2. How much is a one standard deviation increase in days lost due to accidents predicted to change productivity?

3. What is productivity predicted to be in a mine that is neither vertically integrated nor unionized, and capital, days lost, and price are set at their mean values?

4. For each of the following questions formulate a null hypothesis and test it using equation (1).

4.1 Vertical integration and unionization have the same effect on productivity.

4.2 Capital investment is the only determinant of productivity.

4.3 The relationship between prices and productivity is 0.60.

Question 2 - The file Sales.xlxs contains cross-sectional data on record album sales. It includes the following variables:?id = album identifier?sales = total number of albums sold (in millions) radioplay = total number of minutes of radio play songs from the album have received (in 10,000 minutes)?tour = a dummy variable equal to 1 if the band has advertised an album using a tour, 0 otherwise independent = a dummy variable equal to 1 if the album is released by an independent record label You are commissioned by an advertising company to investigate the determinants of album record sales success. Estimate the equation

Sales_i = β₀ + β₁radioplayi + β₂touri + β₃independenti + ε_i                (2)

Using STATA.

1. How many more/less albums are expected to be sold if an album is released by an independent record label?

2. It has been suggested that the effect of radio play on album sales differs depending on whether an album is advertised using a tour. Outline how you would test this and provide a null hypothesis. Do the results support the hypothesis?

3. Using the estimated coefficients from the previous question, how much is a one standard deviation increase in radio play expected to increase/decrease sales if an album is advertised using a tour?

5. Are sales of albums released by independent record labels affected by tours?

6. Test the hypothesis that radio play, tour, and independent have no effect on album sales.

Question 3 - The file BHPS.xlsx contains individual data from the 2014 wave of the British Household Panel Survey. It contains the following variables:?pid = an individual person identifier?female = a dummy variable equal to 1 if a person is female, 0 otherwise?scend = the age at which a person finished education doby = year in which the person was born hourlywage = the hourly wage a person earns (in 's)

1. Using the data contained within BHPS.xls investigate the hypothesis that gender, schooling and age affect earnings. Comment on the results and provide intuition on the economic mechanisms underlying your findings.

2. Test for the presence of heteroskedasticity.

Your answer should be no more than 2000 words.

Attachment:- Assignment Files.rar