You are given a data set with monthly observations. The dataset contains four funds with IDs 1, 21, 38, and 39. Time period of this dataset is from 01/2006 to 12/2013. In other words, 96 fund month observations per fund and 384 observations overall.

Variable Description: (six datasets are posted on D2L)

Ret                              monthly returns

Ttop                            investment in top 10% holdings

Exp                              expense ratio

Turn                            turnover ratio

Holdings                     number of holdings

Tna                              total assets

Date                            Date

Name                          Name

ID                                ID

Additionally, a dataset called FF contains six variables over the same period. Variable Description:

RM                              Market Return

SMB                            the difference in returns between small and large capitalization stocks HML      the difference in returns between high and low book-to-market stocks MOM           the difference in returns between stocks with high and low past returns

RF                               the one-month T-bills rate

Date                            Date

You need to analyze the data and write a thorough report of your analysis. Please be specific; irrelevant work will not earn any points.

Mimic tables 1,2,3,4 and 6 from your reading (Kaushik, A. and A. K. Pennathur (2012) “An Empirical Examination of the Performance of Real Estate Mutual Funds 1990-2008”, Financial Services Review, 2012, Vol. 21 (4), 343-358.).

Important:

Mimic does not mean you need to get the same numbers as reported in the paper, but to use the same template. Remember, the dataset provided to you is not identical to that paper’s dataset.

Specifically,

1.           Annual descriptive statistics for table 1. (HINT: you have monthly observations and you need to prepare annual descriptive statistics: the following command can help you to create a variable that puts together observations belonging to each calendar year:

data X; set X;

annual = Year (date);

run;

(I just called the dataset X as an example, but you can give your major dataset any name…hope you will find some interesting name for that dataset than just naming as X. Painted in yellow is the coding you must apply; again you can call that variable anything. I named it annual, but the command on the right hand side is a must Year (date);)

2.           Run the single factor market model (the CAPM) both overall and per ID.

3.           Re-run the CAPM with three additional factors (SMB, HML, and MOM) as done in table 2 of the paper mentioned above.

4.           Estimate “rolling betas” and report them. Use time interval from 30 months to 96 months for each fund. Do it both ways, the CAPM and the 4-factor Carhart (1997) model.

5.           Estimate rolling monthly 4-factor alphas (HINT: you will have intercept and beta (all 4 betas) coefficients from the 4-factor model. Merge them with the overall dataset by ID and estimate alpha. You have read papers so I am sure you are familiar with the four-factor model. In any case, it is:

(Rit-RFt) = αi + β1*RMRFt + + β1*SMBt+ β1*HMLt+ β1*MOMt+ β1*ɛit

In other words, αi = (Rit-RFt) - β1*RMRFt + β2*SMBt+ β3*HMLt+ β4*MOMt+ ɛit

(HINT HINT…care will be needed when you do merging….)

6.           Regress monthly alphas (from the preceding point (bullet point 5 above)) on cross- sectional variables to complete the cross-sectional analysis. Do overall and by ID.