The McGraw-Hill Series Economics essentials of economics brue, McConnell, and Flynn Essentials of Economics

12.7
What to Do When You Find Autocorrelation:
Remedial Measures
If after applying one or more of the diagnostic tests of autocorrelation discussed in the pre-
vious section, we find that there is autocorrelation, what then? We have four options:
1. Try to find out if the autocorrelation is 
pure autocorrelation
and not the result of
mis-specification of the model. As we discussed in Section 12.1, sometimes we observe
patterns in residuals because the model is mis-specified—that is, it has excluded some
important variables—or because its functional form is incorrect.
34
Ron C. Mittelhammer et al., op. cit., p. 547. Recall that the 
power of a statistical test
is 1 minus
the probability of committing a Type II error, that is, 1 minus the probability of accepting a false
hypothesis. The maximum power of a test is 1 and the minimum is 0. The closer the power of a test
is to zero, the worse is that test, and the closer it is to 1, the more powerful is that test. What these
authors are essentially saying is that there is no single most powerful test of autocorrelation.
guj75772_ch12.qxd 14/08/2008 10:40 AM Page 440

---

Chapter 12
Autocorrelation: What Happens If the Error Terms Are Correlated?
441
2. If it is pure autocorrelation, one can use appropriate transformation of the original
model so that in the transformed model we do not have the problem of (pure) autocorrela-
tion. As in the case of heteroscedasticity, we will have to use some type of 
generalized
least-square (GLS) method.
3. In large samples, we can use the
Newey–West
method to obtain standard errors of
OLS estimators that are corrected for autocorrelation. This method is actually an extension
of White’s heteroscedasticity-consistent standard errors method that we discussed in the
previous chapter.
4. In some situations we can continue to use the OLS method.
Because of the importance of each of these topics, we devote a section to each one.