You must submit a py file that answers these questions in the console. Use the Industry Portfolios.csv as your test assets, converting them to excess returns using RF in FF3.csv. Use real consumption growth (percentage change) and inflation (percentage change of PCEPI) from consumptiondata.csv as your nontradeable factors. Find real consumption (in 2012 dollars) by PCE/(PCEPI/100) (to understand why, go to St. Louis Feds FRED database (where I downloaded the data using pandas datareader) and look up those mnemonics). Use all the data you can.

1. Estimate the general nontradable factor model. For all industries, report the intercept and slope parameters. Using Whites (1980) robust standard errors, report which of these are significant at the 5% level

2. Interpret the results for one of the industry portfolios using financial concepts like risk, exposure, and average return.

3. Use the general nontradable factor model. Use the usual LR statistic and 5% level to test if inflation does not explain the covariation of industry returns. Use the usual LR statistic and 5% level to test if consumption growth does not explain the covariation of industry returns. State if the finite-sample adjusted LR statistic changes your inference.

4. Use the LR statistic and 5% level to test the APT hypothesis. Interpret what this implies about expected returns

5. Given the results in question 3, do you suggest we change the model? If so, how? If not, why?