Use the data in HSEINV1.WFL for this exercise. (1) Let invpc=inv/pop, which means the invest per capita. Let t as the time variable taking 0 with year=1947.(2) Estimate the linear trends for invpc and price separately. And plot the actual time series and its trend together. (3) Regress invpc on price. Write down the results in the usual form. Give your basic analysis. (4) Run unit root test on time series invpc and price to check whether they are stationary. Can you believe the analysis in step (3)?(5) Run unit root test for the first order difference of price. (6) Regress invpc on D(PRICE). Do you think the results are more believable than that of (3)? Why?(7) Remove the linear trend from the PRICE, and then regress the INVPC on the detrended PRICE (denoted as DTPRICE). What will you find?(8) Comparing the regressions in (3) and (8), can you understand the spurious regression relationship? (9) Test whether there is serial correlation in the model of (6). If it has serial correlation, give your correcting on it. (10) While you run the test in (9), do you partial out the effect of exogenous variable(s)?

速度啊