FINM8006 Advance Investments

The purpose of the project is to replicate/extend the Fama and French’s (1993) main findings based on the three stock-market factors, namely (i) an overall market factor (excess market  return) and factors related to (ii) firm size and (iii) book-to-market equity.  Collect the data provided by Professor Kenneth R. French from the following data source: (http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html#Research). The  data library contains the U.S. research returns data, namely monthly “Fama/French 3  Factors” and “25 Portfolios Formed on Size and Book-to-Market (5 x 5)”.
The project steps are as follows:
1. Read the paper “Common risk factors in the returns on stocks and bonds”, Journal  of Financial Economics 33 (1993), 3-56. You can download the paper from here:  http://rady.ucsd.edu/faculty/directory/valkanov/pub/classes/mfe/docs/fama_french_jfe _1993.pdf.
2. The variables required for this project (and discussed in sections 2 and 3 of the paper)
are as follows:
• The Fama/French factors (Mkt-RF, SMB and HML) and the risk-free rate (1- month Treasury bill rate, RF); • 5 x 5 Portfolios Formed on Size and Book-to-Market. The 5x5 portfolios are the dependent variables and the Fama/French factors are the  independent variables. You have to merge two datasets. Use excess stock returns in your estimations.
3. The replication period: July 1963 to December 1991 (N = 342 months). The extension  period: January 1992 to December 2018 (N = 324 months).
4. Replicate the regression results reported in Tables 2 (on p.15 only), 4, 5, 6, and 9a.  Replicate the results with 5x5 stock portfolios. Do not replicate the results with government and corporate bonds. The replication period is from July 1963 to  December 1991.
5. Redo the same analysis as in step (4) in the extension period (January 1992 to  December 2018).
6. Test for January effect in the dependent returns, explanatory returns, and residuals from the three-factor regressions (similar to the test reported in Table 10 but with three factors only). Test for 5x5 stock portfolios only. Test for January effect in the  replication period only.
7. Provide a brief discussion (1-2 paragraphs per question) on the following questions:
• Tables 2, 4, 5, and 6: How do the slope coefficient and R2 estimates change across portfolios and model specifications? Why do they change?
• Table 9a: How do the intercept estimates change across portfolios and model  specifications? Why do they change?
• Tables 2, 4, 5, 6 and 9a: Are the estimates consistent or not in both the replication  and extension periods?
• Table 10: Does January effect hold in the three-factor regressions? Is it consistent  or not across portfolios?
8. Tabulate your estimation results in the same format as reported in the paper  (coefficients, t-statistics, R2 and s(e) ranked on Size and BE/ME quintiles). All  estimations have to be run in STATA.
The magnitudes of your coefficient estimates may be somewhat different from those reported in Fama and French (1993). Submit your assignment report (word or pdf) and all STATA files (.do and .dta files) using  Turnitin in the course WATTLE site. Please ensure I can run all STATA commands in your  .do file and replicate all results in your report. The due date is 10am on Monday, 7th October 2019