TOPIC 4:

ECONOMETRICS
PROBLEMS
Autocorrelation and
Multicollinearity

Autocorrelation
 Autocorrelation
 In situation like this, the assumption of no
auto or serial correlation in the error term
that underlies the CLRM will be violated
 Autocorrelation may be defined as
“correlation between members of series of
observations ordered in time [as in time
series data] or space [as in cross-sectional
data].’’the CLRM assumes that:E(uiuj)=0 for
i ≠ j, but in this scenario E(uiuj )≠0 for i ≠ j.
Friday, July 16, 2021 3
 Example 1: if we are dealing with quarterly
time series data involving the regression of
output on labor and capital inputs and if, say,
there is a labor strike affecting output in one
quarter, there is no reason to believe that this
disruption will be carried over to the next
quarter. That is, if output is lower this quarter,
there is no reason to expect it to be lower
next quarter.
Friday, July 16, 2021 4
 Example 2: If we are dealing with cross-
sectional data involving the regression of
family consumption expenditure on family
income, the effect of an increase of one
family’s income on its consumption
expenditure is not expected to affect the
consumption expenditure of another family

Friday, July 16, 2021 5

 The causes of autocorrelation
1:Omitting an important variable, that is one
variable is affected by so many variable, The
error term represent the influence of omitted
variable and because of that an error term in
one period many have relation with the error
term in successive periods. Thus the period
of autocorrelation occurs.Yt=β1+β2X22t+ut but
we wrongfully assume and estimate a
straight line:Yt=β1+β2X2t+ut Then the error
term obtained from the straight line will
depend on X22t
Friday, July 16, 2021 6
2:Inertia.A salient feature of most
economic time series is inertia, or
sluggishness. As is well known, time series
such as GNP, price indexes, production,
employment, and unemployment
exhibit(business) cycles.

Friday, July 16, 2021 7

3: Specification Bias- Incorrect Functional
Form
Yt   1   2 X 2 t   3 X 22t  vt

Yt   1   2 X 2 t  u t

u t   3 X 22t  v t

Friday, July 16, 2021 8

4:Manipulation’’of Data.In empirical analysis,
the raw data are often “manipulated.’’ For
example,in time series regressions involving
quarterly data, such data are usually
derived from the monthly data by simply
adding three monthly observations and
dividing the sum by 3.This averaging
introduces smoothness into the data by
dampening the fluctuations in the monthly
data..
Friday, July 16, 2021 9
. 5: Nonstationarity. A time series is stationary
if its characteristics (mean, variance, and
covariance) they do not change over time. If
that is not the case, we have a nonstationary
time series. In a regression model such as: Yt
= β1 + β2Xt + ut ;it is quite possible that both Y
and X are nonstationary and therefore the
error u is also nonstationary. In that case, the
error term will exhibit autocorrelation.
Friday, July 16, 2021 10
6: Cobweb Phenomenon. The supply of
many agricultural commodities reflects the
so-called cobweb phenomenon, where
supply reacts to price with a lag of one time
period because supply decisions take time
to implement (the gestation period). Thus,
at the beginning of this year’s planting of
crops, farmers are influenced by the price
prevailing last year.
Supplyt = β1 + β2Pt−1 + ut
Friday, July 16, 2021 11
7: Lags
Consumption t   1   2 Consumption t 1  u t
 The above equation is known as autoregression
because one of the explanatory variables is the
lagged value of the dependent variable.
 If you neglect the lagged the resulting error term
will reflect a systematic pattern due to the
influence of lagged consumption on current
consumption.
Friday, July 16, 2021 12
 Consequences of
autocorrelation
1:When disturbance term is seriously correlated
then the least (OLS) estimation is unbiased
but optimistic property (minimum variance
property) is not satisfied.
2: If disturbance term is autocorrelated then the
OLS variance is greater than the variance of
estimated calculated by the other methods
therefore the use of i and f of significant are no
longer valid.
Friday, July 16, 2021 13
3. The regression coefficient remains
unbiased but OLS is inefficient because one
can alternate unbiased estimated estimator
with the smaller variance.
4. The variance of random tem u may be
seriously underestimated if the u’s are
autocorrelated.
5. If the disturbance term are auotocorrelated
then the ordinary least square (OLS) estimate
are not asymptotic.
Friday, July 16, 2021 14
6.OLS estimated is no longer blue, GLS is
blue
7. Estimators are still linear and unbiased but
they are not the best, they are inefficient

Friday, July 16, 2021 15

 Ways to detect autocorrelation

1.Graph
2.The Durbin Watson Test
Step 1: Estimate the model by OLS and obtain the
residuals
Step 2: Calculate the DW statistic
Step 3: Construct the table with the calculated DW
statistic and the dU, dL, 4-dU and 4-dL critical
values.
Step 4: Conclude
Friday, July 16, 2021 16
3.The Breusch-Godfrey Test
4.The Engle’s ARCH Test

Friday, July 16, 2021 17

 REMEDIAL MEASURES
1. Try to find out if the autocorrelation is
pure autocorrelation and not the result
of mis-specification of the model.
sometimes we observe patterns in
residuals because the model is
misspecified—that is, it has excluded
some important variables—or because its
functional form is incorrect.
Friday, July 16, 2021 18
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
autocorrelation. As in the case of
heteroscedasticity, we will have to use
some type of generalized least-square
(GLS) method.

Friday, July 16, 2021 19

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.
4. In some situations we can continue to
use the OLS method.

Friday, July 16, 2021 20

Multicollinearity
Introduction
 If Xi and Xj are statistically independent,
then a change in Xi has no correlation to
a change in Xj.
 Usually, however, there is some amount
of correlation between variables.
 Multicollinearity occurs when Xi and Xj
are related to each other.

Friday, July 16, 2021 22

 Sources of Multicollinearity
 1. The data collection method employed,
for example, sampling over a limited range
of the values taken by the regressors in the
population.
 2. Constraints on the model or in the
population being sampled. For example,
in the regression of electricity consumption
on income (X2) and house size (X3) (High
X2 always mean high X3).
Friday, July 16, 2021 23
 3. Model specification, for example,
adding polynomial terms to a
regression model, especially when the
range of the X variable is small.
 4. An overdetermined model. This
happens when the model has more
explanatory variables than the number
of observations.
Friday, July 16, 2021 24
 5. An additional reason for
multicollinearity, especially in time
series data, may be that the
regressors included in the model share
a common trend, that is, they all
increase or decrease over time.

Friday, July 16, 2021 25

 Consequences of
Multicollinearity
 1. Although BLUE, the OLS estimators
have large variances and covariances,
making precise estimation difficult.
 2. Because of consequence 1, the
confidence intervals tend to be much wider,
leading to the acceptance of the “zero null
hypothesis” (i.e., the true population
coefficient is zero) more readily.

Friday, July 16, 2021 26

 3. Also because of consequence 1, the t
ratio of one or more coefficients tends to
be statistically insignificant.
 4. Although the t ratio of one or more
coefficients is statistically insignificant, R 2
can be very high.
 5. The OLS estimators and their standard
errors can be sensitive to small changes
in the data.
Friday, July 16, 2021 27
Detection of Multicollinearity
 How does one know that collinearity is
present in any given situation,
especially in models involving more
than two explanatory variables? Here it
is useful to bear in mind Kmenta’s
warning:

Friday, July 16, 2021 28

 1. Multicollinearity is a question of degree
and not of kind. The meaningful distinction is
not between the presence and the absence
of multicollinearity, but between its various
degrees.
 2. Multicollinearity is a feature of the sample
and not of the population. Therefore, we do
not “test for multicollinearity” but we measure
its degree in any particular sample.
Friday, July 16, 2021 29
 We do not have one unique method of
detecting it or measuring its strength.
What we have are some rules of
thumb, some informal and some
formal.
 The following are the ways to detect
multicollinearity

Friday, July 16, 2021 30

 1.High R2 but few significant t ratios. If R2
is high, say, in excess of 0.8, the F test in
most cases will reject the hypothesis that
the partial slope coefficients are
simultaneously equal to zero, but the
individual t tests will show that none or
very few of the partial slope coefficients
are statistically different from zero.
Friday, July 16, 2021 31
 2. Estimates are very sensitive to changes in
data
 3. Unexpected signs on parameter estimates
and/or unreasonable magnitudes
 4. Significant joint effects for the affected
variables, but insignificant t – tests for the
individual variables.
 5. High correlations between two or more of
the affected variables.
Friday, July 16, 2021 32
 When this happens, there is an
“overlap” between what Xi explains
about Y and what Xj explains about Y.
This makes it difficult to determine the
true relationship between Xi and Y,
and Xj and Y

Friday, July 16, 2021 33

Remedies for
Multicollinearity
What can be done if multicollinearity is
serious? We have two choices: (1) do
nothing or (2) follow some rules of
thumb
The “do nothing” school of thought is
expressed by Blanchard as
Multicollinearity is God’s will, not a
problem with OLS or statistical
technique in general.
Friday, July 16, 2021 34
What Blanchard is saying is that
multicollinearity is essentially a data
deficiency problem (micronumerosity,
again) and some times we have no
choice over the data we have available
for empirical analysis.

Friday, July 16, 2021 35

Following the rules of thumb the
following are the remedies:
 Respecification of the model.
 Determine if the model is correctly
specified.
 Combine the variables by multiplying or
dividing them.

Friday, July 16, 2021 36

 Additional or new data. Since
multicollinearity is a sample feature, it is
possible that in another sample
involving the same variables collinearity
may not be so serious as in the first
sample. Sometimes simply increasing
the size of the sample (if possible) may
attenuate the collinearity problem.

Friday, July 16, 2021 37

 Transformation of variables: by first
difference.
 Dropping a variable(s) and specification
bias.
When faced with severe
multicollinearity, one of the “simplest”
things to do is to drop one of the
collinear variables
Friday, July 16, 2021 38
 Combining cross-sectional and time
series data. A variant of the
extraneous or a priori information
technique is the combination of cross-
sectional and time-series data, known
as pooling the data

Friday, July 16, 2021 39

 NB: Rule of Thumb for determining if
you have multicollinearity:
 Widely varying coefficients
 Correlation Matrix:
 r  0.3 No Problem
 0.3  r  0.7 Gray Area
 r  0.7 Problems Exist

Friday, July 16, 2021 40

 Is multicollinearity necessarily bad?
Maybe not if the objective is
prediction only
 It has been said that if the sole purpose of
regression analysis is prediction or
forecasting, then multicollinearity is not a
serious problem because the higher the R2,
the better the prediction.
 Moreover, if the objective of the analysis is not
only prediction but also reliable estimation of
the parameters, serious multicollinearity will be
a problem because we have seen that it leads
to large standard errors of the estimators.
Friday, July 16, 2021 41
 Thank you for you Attention

Monday, April 27, 2015 42