Autocorrelation:

Nature and Detection

13.1
 Aims and Learning Objectives

By the end of this session students should be able to:

• Explain the nature of autocorrelation

• Understand the causes and consequences of

autocorrelation

• Perform tests to determine whether a regression

model has autocorrelated disturbances

13.2
 Nature of Autocorrelation

Autocorrelation is a systematic pattern in the

errors that can be either attracting (positive)
or repelling (negative) autocorrelation.

For efficiency (accurate estimation/prediction)

all systematic information needs to be incor-
porated into the regression model.

13.3
 Regression Model

Yt = 1 + 2X2t + 3X3t + Ut

No autocorrelation: Cov (Ui, Uj)

or E(Ui, Uj) = 0

Autocorrelation: Cov (Ui, Uj)  0

or E(Ui, Uj)  0
Note: i  j
In general
E(Ut, Ut-s)  0
13.4
 Ut Attracting
Postive
. . .. . . . ..
.. .
Auto. 0
. . .. . . ...
. t

Ut Random
No . . .. . . . . . . . .
Auto. 0
. . .. . . . . .. .
. . . . .t
. Repelling
Ut . . . . .
Negative
0
. . .
Auto.
. . . . . . . t
. . 13.5
 Order of Autocorrelation
Yt = 1 + 2X2t + 3X3t + Ut

1st Order: Ut = Ut1 + t

2nd Order: Ut = 1 Ut1 + 2 Ut2 + t
3rd Order: Ut = 1 Ut1 + 2 Ut2 + 3 Ut3 + t
Where -1 <  < +1
We will assume First Order Autocorrelation:
AR(1) : Ut = Ut1 + t
13.6
 Causes of Autocorrelation

Direct Indirect

• Inertia or persistence • Omitted Variables

• Spatial correlation • Functional form

• Cyclical Influences • Seasonality

13.7
 Consequences of Autocorrelation

1. Ordinary least squares still linear and

unbiased.

2. Ordinary least squares not efficient.

3. Usual formulas give incorrect standard

errors for least squares.

4. Confidence intervals and hypothesis tests

based on usual standard errors are wrong.13.8
 ^ ^
Yt = 1 + 2Xt + et
Autocorrelated disturbances: E(et, et-s)  0
Formula for ordinary least squares variance
(no autocorrelation in disturbances): ˆ  2
Var (  2 )  2
x t

Formula for ordinary least squares variance

(autocorrelated disturbances): 2
  
Var ( ˆ 2 )  1  1 2 xi x j  k 
 t  t
x 2 
x 2 


Therefore when errors are autocorrelated ordinary

least squares estimators are inefficient (i.e. not “best”)
13.9
 Detecting Autocorrelation

Yt  ˆ1  ˆ2 X 2t  ˆ3 X 3t  et

et provide proxies for Ut

Preliminary Analysis (Informal Tests)

• Data - autocorrelation often occurs in time-series

(exceptions: spatial correlation, panel data)

• Graphical examination of residuals - plot et against

time or et-1 to see if there is a relation
13.10
 Formal Tests for Autocorrelation

Runs Test: analyse the uninterrupted sequence of the

residuals

Durbin-Watson (DW) d test: ratio of the sum of

squared differences in successive residuals to the
residual sum of squares

Breusch-Godfrey LM test: A more general test

which does not assume the disturbances are AR(1).

13.11
 Durbin-Watson d Test

H o:  = 0 vs. H1:  = 0 ,  > 0, or  < 0

The Durbin-Watson Test statistic, d, is :
n
et et-1
2
t=2
d = n
et
t=1
2

Ratio of the sum of squared differences in successive

residuals to the residual sum of squares
13.12
The test statistic, d, is approximately related to ^
 as:

d  2(1)
^

 = 0 , the Durbin-Watson statistic is d  2.

When ^

 = 1 , the Durbin-Watson statistic is d  0.

When ^

 = -1 , the Durbin-Watson statistic is d  4.

When ^

13.13
 DW d Test
4 Steps

Step 1: Estimate Yˆi  ˆ1  ˆ2 X 2i  ˆ3 X 3i

And obtain the residuals
Step 2: Compute the DW d test statistic

Step 3: Obtain dL and dU: the lower and upper points

from the Durbin-Watson tables

13.14
 Step 4: Implement the following decision rule:
Value of d relative to dL and dU Decision

d < dL Reject null of no positive

autocorrelation
dL  d  dU No decision

dU < d < 4 - dU Do not reject null of no

positive or negative
autocorrelation

4 – dL < d < 4 - dU No decision

d > 4 - dL Reject null of no negative

autocorrelation

13.15
 Restrictive Assumptions:
• There is an intercept in the model

• X values are non-stochastic

• Disturbances are AR(1)

• Model does not include a lagged dependent

variable as an explanatory variable, e.g.

Yt = 1 + 2X2t + 3X3t + 4Yt-1+ Ut

13.16
 Breusch-Godfrey LM Test

This test is valid with lagged dependent variables

and can be used to test for higher order
autocorrelation

Suppose, for example, that we estimate:

Yt = 1 + 2X2t + 3X3t + 4Yt-1+ Ut

And wish to test for autocorrelation of the form:

U t  1U t 1   2U t 2  3U t 3  vt
13.17
 Breusch-Godfrey LM Test
4 steps

Step 1. Estimate
Yt = 1 + 2X2t + 3X3t + 4Yt-1+ Ut
obtain the residuals (et)

Step 2. Estimate the following auxiliary regression

model:
et  b1  b2 X 2  b3 X 3  b4Yt 1
 c1et 1  c2et 2  c3et 3  wt 13.18
 Breusch-Godfrey LM Test

Step 3. For large sample sizes, the test statistic is:

( n  p) R ~ 
2 2
p

Step 4. If the test statistic exceeds the critical

chi-square value we can reject the null hypothesis
of no serial correlation in any of the  terms

13.19
 Summary

In this lecture we have:

1. Analysed the theoretical causes and

consequences of autocorrelation

2. Described a number of methods for detecting

the presence of autocorrelation

13.20