Professional Documents
Culture Documents
1 Autocorrelated errors
Here we are now considering what happens when this is no longer the case, i.e
when cov( i , j ) = E ( i j ) 0 for all i j .
We need to consider:
1
Autocorrelated variables:
X t = X t −1 + t
2
Let’s first look at the variance of this variable:
Note: this proof has assumed that X t is homoskedastic, i.e. the variance is the
same no matter what time period we are in ( var ( X ) = var ( X ) ). t t −1
Now let’s look at the covariance between the variable and its first lag:
cov ( X t , X t −1 ) = E ( X t X t −1 ) = E ( ( X t −1 + t ) X t −1 ) = E ( X t2−1 + t X t −1 )
3
We can also see that there is a non-zero second-order autocorrelation:
(
cov ( X t , X t −2 ) = E ( X t X t −2 ) = E ( ( X t −1 + t ) X t −2 ) = E ( ( X t −2 + t −1 ) + t ) X t −2 )
= E ( X 2 2
+ t −1 X t −2 + t X t −2 ) = 2 2
t −2
(1− )
2
and in general
cov ( X t , X t − j ) = E ( X t X t − j ) = j 2
(1− ) .
2
Given that the autocorrelation coefficient is less than 1 in absolute value, then
j → 0 as j → . Hence as we look at autocorrelations further and further into the
past, the correlation gets smaller. This makes sense - a variable is more likely
to be correlated with itself in the near past than the distant past.
4
The following are plots of a simulated AR(1) process X that exhibits no
autocorrelation ( = 0 ), positive autocorrelation ( 0 ) and negative
autocorrelation ( 0 ).
When = 0 , the plot shows a fairly random pattern with no discernible pattern
or predictability.
5
When = 0.9 there is strong positive autocorrelation. Here, the pattern is
smoother with runs of positive and negative values and looks a bit like a cycle.
6
When = −0.9 there is strong negative autocorrelation. Negative
autocorrelation shows a spikier plot in which a positive value in one period is
likely to be followed by a negative value in the next period.
7
This sub-section has shown you what autocorrelated variables in general look
like.
We are concerned here with autocorrelation that arises in the error terms of
our regression model, i.e if the disturbance from a regression model is
autocorrelated.