(You can submit/upload the complete ipynb file in the canvas)
 
TASK A. DATA VISUALIZATION: 4 POINTS
Get the data “30_Industry_Portfolios.csv” data and “F-F_Research_Data_Factors.CSV” from class google drive.
For both data select date from January 01, 2000 to December 31, 2018
Draw a 2*2 scatterplot that reflect the relationship between (1) mkt-rf and food, (2) SMB and food, (3) mkt-rf and games and (4) SMB and games.
From the scatterplot what kind of relationshiop you see. [explain in words]
TASK B. REGRESSION SPECIFICATION- USE THE OLS FUNCTION TO ESTIMATE THE FOLLWOING: 10 POINTS
Run two univariate regression on Food and Games using excess market return (mkt-rf) as the dependent variable. Include a constant in the regressions.
Report the coefficient and t-stat of the market in this both industry, are they significant?
Report the R-squares of the models.
Run a multivarite regression for this two industry return (Food and Games) using all three Fama French Factors ( Market, SMB and HML).
Report the coefficient and t-stat of all three factors in both models, are they significant?
Report the R-squares of the models.
Is their any difference in the R-squares on 3 and 6. What doese this mean?
TASK C. WHITE’S COVARIANCE ESTIMATOR: 6 POINTS
Re-estimate the two multivarite regression (Task B.4) with White’s covariance estimator.
Report the coefficient and t-stat of all three factors in both models, are they significant?
Are the parameter standard errors similar using the two covariance estimators (homoskedastic errors in Task B.4 and White’s covariance estimator). If not, what does this mean?