Reference no: EM132264002
PROJECT - ESTIMATION OF CAPM AND THE FAMA-FRENCH THREE-FACTOR MODEL
Part I: CAPM
1. Request a password for WRDS (Wharton Research Data Service) database.
2. In WRDS, find Compustat database. In Compustat, you can find information on the book value and market value of equity, and market capitalization.
3. Based on the above information, select two companies that you will work with throughout this project. The first company should be large capitalization stock with high book to market (BM) ratio. The second company should be a small cap stock with low BM ratio. You can check how Prof. Kenneth French constructs these ratios. Look at the explanations from the data library on his website.
Report the information on book value, market value, etc., for your two companies.
4. In WRDS find the database, called CRSP. It contains security prices, returns and volume data of NYSE, AMEX, and NASDAQ firms. You can also use Bloomberg.
5. Download monthly returns for the most recent five-year period for your two companies. Also, download monthly return data for the same time period for the S&P 500 index (or any other large value-weighted index, which you intend to use as a proxy for the market portfolio). Pace does not have a subscription to the Indices database. To circumvent this problem, use the option to download "Returns relative to the S&P 500 index". You will get returns on S&P 500 next to the returns on the chosen stock.
6. Download risk-free rate data (returns on 3-month T-bills) for the same time period.
Choose rates on T-bills (the section is called T-bills) with 3-months to maturity. Download data at monthly frequency. Then calculate the average monthly risk free rate for your chosen five-year time period. Use AVERAGE function. Note, Fed reports annualized returns, i.e. monthly returns are already multiplied by 12.
7. Estimate betas for your two companies via the regression
rj - rf = αj + βj (rm - rf) + εj
for each of your two chosen stocks. In Excel:
a. Go to Tools.
b. Go to Data Analysis.
c. Choose Regression. Your Y variable is (monthly return on your chosen stock - the average rf) and your X variable is (monthly return on S&P 500 - the average rf).
8. Report your estimate of beta and the standard error of beta, σβ, from both available in the regression output. Your report should contain the Excel output from your regressions in the Appendix.
9. Construct 95% confidence intervals for your estimates of beta (see GE example in the notes on the Empirical Issues and CAPM). Compare confidence intervals from both regressions and comment on the difference in your results.
10. Report R2 (coefficient of determination) from both regressions. What does R2 mean? Comment on the difference between the coefficients of determination from two regressions.
11. Test a hypothesis that α = 0 (since you run a regression through the origin) vs. the alternative that α ≠ 0. It is a double sided t test.
12. Comment on your results: How well CAPM explains returns on large-cap value stocks and on small-cap growth stocks.
PART II: FACTOR MODELS
1. From Prof. French's website download data on the market index (SP 500) as well as SMB and HML factors at monthly frequencies for the same period of time as in part 1 of this project.
2. Estimate the Fama-French Three Factor Model for each of your two stocks. You are estimating coefficients of the multiple regression with three independent (X) variables:
R~it - rf = α^i + β^Mi(R~Mt - rf) + β^SMBi SMBt + β^HMLi HMLt + εit
3. Report the results of your estimation and comment on their statistical significance:
a. Which beta coefficient are significant at the 5% confidence level (use t statistic).
b. What are the coefficients of determination and what do they mean?
4. Compare results from the two regressions:
a. Report coefficients of determination and comment on the difference between two coefficients from the regressions on your two stocks.
b. Compare individual coefficients from both regressions.
5. Compare and contrast results from the CAMP model and FF models for the large cap/value stock. Which model seems to explain and predict returns better? Comment, using results of your regression analysis.
6. Does FF model outperform CAPM in explaining returns on the small cap/growth stock?
Attachment:- Assignment File.rar