10+10+5+5+10=40 points) Consumer Price Index (CPI, base year 1982-84=100) and the New York Stock Exchange Index (NYSE) (base year, 1965=100) for the period 1977-1991. Using Stata, run two simple regressions goldpricet = + CPIt + ut NY SEIndext = + CPIt + ut

1. Estimate these two simple regressions and attach stata output to your home- work.

2. State the hypothesis that CPI has no effect on goldprice i.e. test for the individual significance based on a two-tailed t-test. Carry out the test at = 10%. Redo the same exercise for the effect of CPI on NYSEIndex. Does CPI have an individually significant effect on goldprice? What is the p-value for this t-statistic? Is it less than = 10%? Does CPI have an individually significant effect on NYSEIndex ?

3. What is the predicted increase in goldprice given a unit increase in CPI ? What is the predicted increase in NYSEIndex given a unit increase in CPI ? In which of the two models, R2 is higher? Give a verbal interpretation of this statistic in the context of the model.

4. Now calculate the correlation coefficient, rxy using Stata command, corr, for these two variables, NYSEIndex and CPI. Verify that R2 reported in your regression output is equal to r2xy (only in simple regression).

5. n investment is supposed to be a good hedge against inflation if its price and/or rate of return at least keeps pace with inflation. To test this hypothesis, what value of would you expect? Based on your estimates of in the two regressions you fit, which is a better hedge against inflation, gold or the stock market? Be clear and justify your answers based on the statistics you observe in the output.