Questions -
Q1. a) Using a t-test and a 5% level of statistical significance, are the individual estimated effects, other than the constant term, each statistically significant from zero.
b) Test the regression for first-order autocorrelation by using Durbin Watson d-statistic.
c) Using a t-test, does the coefficient β equal one?
d) Test for 2nd order autocorrelation using the LM test for autocorrelation.
e) In the spirit of arbitrage pricing theory (APT), the following example will examine regressions that seek to determine whether the monthly returns on Microsoft stock can be explained by reference to unexpected changes in a set of marcoeconomic and financial variables. Use the data file 'macro.wf1' containing 326 monthly observations from March 1996 to April 2013. The database contains following series: Microsoft stock price, S&P500 index value, consumer price index, industrial production index, Treasury bill yields for following maturities (three months, six months, one year, three years, five years and then years), a measure of narrow money supply, consumer credit series, and a credit spread series. The later is defined as the difference in annualised average yields between a portfolio of bonds rated AAA and a portfolio of bonds rated BAA.
Q2. a) Compute the first difference for each of the variables. Give some intuition behind the differences.
b) Estimate the impct of financial and marcovariables on the Microsoft exces returns with OLS. Go to Quick than click on option Estimate Equation: Open new window right the command (ermsoft c ersandp dprop dcredit dinflation dmoney dspread rterm). Are the coefficeints are statistically significant?
c) Do regression is jointly significant?
d) Test the joint level of significance for the coefficients DPROD, DCREDIT, DINFLATION, DMONEY and DSPREAD.