The file SP500TBIBM contains 1008 daily observations on three months T-bill rates, and daily closing prices of the SP 500 index and IBM stock...

The file SP500TBIBM contains 1008 daily observations on three months T-bill rates, and daily closing prices of the S&P 500 index and IBM stock between November 1, 1993 to March 31, 1999.

Use the code in file Assgn2.sas to calculate the excess returns on S&P 500 and IBM and estimate the CAPM model, as shown on page 239 of the textbook.

Do you find a statistically significant correlation between the excess returns on IBM and excess market returns?

Are the results of the estimation without an intercept consistent with those when the intercept is included?

What is the test of the null hypothesis "intercept = 0" ?

Do the data provide sufficient evidence that the CAPM holds?

SAS Code

options linesize=78;* the data *;data capm;infile ;input tb sp ibm;* compute returns on IBM *;retibm = dif((ibm));t=_n_;* compute returns on SP500 *;retsp = dif((sp));* transform T-bill (page 239) *;tbil = tb/25300;* calculate excess returns *;exretsp = retsp - tbil;exretibm = retibm - tbil;* estimate the CAPM model (page 239) *;proc reg;model exretibm = exretsp;run; * estimate the CAPM model no intercept (page 240) *;proc reg;model exretibm = exretsp/noint;run;* 10 first datapoints *;data t10; capm; t le 10;run;proc data=t10;run;

SP500TBIBM