Heteroscedasticity Outputs -Need tomorrow (Sat)

Autocorrelation Problem Set

Refer to the attached “stocks” data. Data include NYSE index values, GDP measured in billions of dollars, and time from 1980-2006. First, estimate the following equation using Ordinary Least Squares (OLS):

Add detailed content

a. Use a scatter diagram to assess whether there is first-order autocorrelation in this series?

Add detailed content

b. Now, assess whether there is first-order autocorrelation based on the basis of the d test. Note: use the command “tsset time” so that you can estimate the Durbin-Watson statistic using “estat dwatson.”

Durbin-Watson d statistic

Add detailed content

Add detailed content

c. Finally, correct the problem of autocorrelation using the “Newey-West” method.

d. (Optional) Use the d value from (b) to transform the data per the generalized difference equation below:

Is there autocorrelation in the transformed model?

Add detailed content

-1000

-500

0

500

1000

1500

Residuals

198019851990199520002005

time

-1000

-500

0

500

1000

1500

Residuals

-1000-500050010001500

Residuals, L