Scribd
Upload
152 views10 pages

Topic6 Autocorrelation PDF

This document discusses autocorrelation in time series data. It begins by defining autocorrelation as the correlation of observations across time periods. It then explains that autocorrelation violates the assumption of uncorrelated errors in ordinary least squares regression. The document outlines some common causes of autocorrelation and shows that while OLS remains unbiased in the presence of autocorrelation, it is no longer efficient. The best linear unbiased estimator in the case of autocorrelation is generalized least squares which accounts for the autocorrelated errors.

Uploaded by

Wong Wei Xiang Gerald
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
152 views10 pages

Topic6 Autocorrelation PDF

This document discusses autocorrelation in time series data. It begins by defining autocorrelation as the correlation of observations across time periods. It then explains that autocorrelation violates the assumption of uncorrelated errors in ordinary least squares regression. The document outlines some common causes of autocorrelation and shows that while OLS remains unbiased in the presence of autocorrelation, it is no longer efficient. The best linear unbiased estimator in the case of autocorrelation is generalized least squares which accounts for the autocorrelated errors.

Uploaded by

Wong Wei Xiang Gerald
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

HE2004/HE204A Introductory Econometrics

Semester 1, 2013 Autocorrelation

September 2013

HE2004/HE204A Introductory Econometrics ()

Autocorrelation

September 2013

1/9

Introduction

When dealing with time series data, observations are often correlated across time. Violates the CLR assumption of uncorrelated errors. In this section we look at
1 2 3 4

the nature of autocorrelation consequences of autocorrelation detection of autocorrelation remedial measures.

HE2004/HE204A Introductory Econometrics ()

Autocorrelation

September 2013

2/9

Nature of autocorrelation
Observations in dierent time periods are correlated, i.e. E (ui uj ) 6= 0 for i 6= j . "Autocorrelation" and "serial correlation" Reasons for autocorrelation:
1 2 3 4 5 6 7

Graphically, "stickiness" of the errors (estimated) suggest correlation.

Most economic series exhibit business cycles. Omitted variables. Incorrect functional form. Cobweb phenomenon. Data manipulation e.g. smoothing, interpolation, extrapolation Transformation of model e.g. rst dierencing. Non-stationarity in dependent and independent variables.
Autocorrelation September 2013 3/9

HE2004/HE204A Introductory Econometrics ()

What implication is autocorrelation for OLS estimation? Consider the two-variable model:

[6.1] [6.2]

Yt = 1 + 2 Xt + u t ut = ut
1

+ t ,

1<<1

t satises the standard OLS assumptions:

[6.3] [6.4] [6.5]

E ( t ) = 0 var (t ) = 2 cov (t , t +s ) = 0, s 6= 0

An error term with the preceding properties is often referred to as a white noise. ut is said to follow a rst-order autoregressive process, or AR(1) process. For 1 < < 1, the error ut is stationary, i.e. constant mean, variance and covariances depend on the time interval between the errors.
HE2004/HE204A Introductory Econometrics () Autocorrelation September 2013 4/9

It can be shown that

[6.6] [6.7] [6.8] [6.9]

E ( ut ) = 0
2) = var (ut ) = E (ut 2 1 2
2

cov (ut , ut +s ) = E (ut , ut +s ) = s 1 2 , cor (ut , ut +s ) = s

s = 0, 1, 2, ....

Under the standard OLS assumptions, the OLS estimator of 2 is

[6.10]

where xi = Xi

b 2 =

x i y i xi2

X and yi = Yi
2 xi2

Y , and its variance is

[6.11]

HE2004/HE204A Introductory Econometrics ()

var (b 2 ) =

Autocorrelation

September 2013

4/9

Under AR(1) errors, however, the variance of [6.10] is

[6.12]

2 xt2

var (b 2 )AR 1 =

t xt 1 + 2 x x2 t

t xt 2 + 22 x + x2 t

+ 2n

1 22 x 1 x n xt2

Note the [6.12] is not the same as [6.11], unless = 0. So, usual OLS variance is a biased estimator of the true variance. The true variance [6.12] depends on the sample correlation between the X values, so it is sample dependent. In general, it is not possible to tell if var (b 2 ) > var (b 2 )AR 1 or var (b 2 ) < var (b 2 )AR 1 . Hence, inference based on [6.11] would be incorrect. What about the linear unbiased property of OLS? Under autocorrelation OLS estimator continues to be linear and unbiased. But it is not "best", ie. it is not e cient. Within the class of linear unbiased estimators, there is a more e cient estimator.
HE2004/HE204A Introductory Econometrics () Autocorrelation September 2013 5/9

BLUE under Autocorrelation


Lag [6.1] by 1 period and multiply by to get

[6.13]

Yt

= 1 + 2 Xt

+ ut

Subtract [6.13] from [6.1]:

[6.14]

Yt

Yt ut

1 1

= (1

) 1 + 2 ( Xt

Xt

1 ) + vt

where vt = ut Note that

[6.16]

E ( v t ) = E ( ut

ut

1)

=0

var (vt ) = var (t ) = 2


HE2004/HE204A Introductory Econometrics () Autocorrelation September 2013 6/9

BLUE of 2 is obtained by OLS on [6.14]

[6.17] [6.18]

Equation [6.17] is not feasible since usually we do not know . A feasible version of it involves estimating which we will discuss later.

n t =2 (x t x t 1 )(y t y t 2 n t =2 (x t x t 1 ) GLS 2 var (b 2 ) = n (xt xt 1 )2 t =2

GLS b 2 =

1)

HE2004/HE204A Introductory Econometrics ()

Autocorrelation

September 2013

7/9

Consequences of OLS in Autocorrelation


Under autocorrelation, OLS continues to be linear, unbiased, consistent and asymptotically normal, if the other standard assumptions hold. But it is no longer e cient. What about the usual hypothesis testing procedures? Hypotheses testing based on var (b ) is clearly invalid.
2

2 / (n b 2 = u The residual variance bt 2) is likely to underestimate the 2 true variance . When there is AR(1) in the error term

[6.19]

1 n 2 where r = n t =1 xt xt 1 / t =1 xt , which can be interpreted as the sample correlation coe cient between successive values of the X 0 s .
HE2004/HE204A Introductory Econometrics () Autocorrelation September 2013 8/9

b2 ) = E (

2 fn [2 /(1 )] 2 r g n 2

Another source of downward biasedness of the estimate of var (b 2 ) can be seen by comparing [6.11] and [6.12]. If > 0 and the X 0 s are positively correlated, then var (b 2 ) < var (b 2 )AR 1 . Hypotheses testing based on var (b 2 )AR 1 , while valid, generally lacks power.

b2 ) < 2 , i.e. the usual residual If both > 0 and r > 0, then E ( variance will underestimate the true variance. This will then lead to downward bias of the estimate of var (b 2 ) since we typically estimate 2 2 b /xt . it by

HE2004/HE204A Introductory Econometrics ()

Autocorrelation

September 2013

9/9

You might also like