Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 23

Heteroskedasticity

Vs.
Homoskedasticity

Chapter no 11
Heteroskedasticity: Definition
• Heteroskedasticity is a problem where the
error terms do not have a constant variance.

E (ei 2 )   i 2

• That is, they may have a larger variance when


values of some Xi (or the Yi’s themselves) are
large (or small).

May 25, 2019 Slide 2


Heteroskedasticity
Homoscadesticity
Heteroskedasticity: Definition
• This often gives the plots of the residuals by
the dependent variable or appropriate
independent variables a characteristic fan or
funnel shape.

180
160
140
120
100
Series1
80
60
40
20
0
0 50 100 150

May 25, 2019 Slide 5


Reasons of Heteroscedasticity:
• Following the error learning models, as people
learn, their errors of behavior become smaller
over time. For example, number of typing error.
As the number of hours put in typing practice
increases, the average number of typing errors as
well as their variances decreases.
• As income grow, people have more discretionary
income and hence more scope for choice about
the disposition of their income. Hence is likely to
increase with income.
• Improvement of data collecting technique cases,
become decreases.
• In the presence of outlier heteroscedasticity
arises.
• Due to incorrect specification of regression
model, it arises.
• If also arises because of incorrect
transformation and incorrect functional form.
• For skewness of distribution of regression, it
arises.
Consequences of Using OLS in the
Presence of Heteroscedasticity:
• The consequences of OLS in the presence of
heteroscedasticity are:
• In the presence of heteroscedasticity, still unbiased,
linear, consistent but inefficient.
• The OLS estimate is biased and inconsistent i.e. .
• The confidence interval based on OLS estimator will be
unnecessary large.
• Variance of OLS coefficients will be incorrect i.e. . In
this case we have to estimate -variance from -
observations. That is, one for each variance, a situation
in which estimation is obviously impossible because we
cannot estimate a variance from on observation.
Heteroskedasticity: Implications
(cont.)

• The estimator variances are not


asymptotically efficient, and they are biased.
– So confidence intervals are invalid.
– What do we know about the bias of the
variance?
– If Yi is positively correlated with ei, bias is
negative - (hence t values will be too large.)
– With positive bias many t's too small.

May 25, 2019 Slide 12


Heteroskedasticity: Implications
(cont.)
• Types of Heteroskedasticity
– There are a number of types of heteroskedasticity.
• Additive
• Multiplicative
• ARCH (Autoregressive conditional heteroskedastic) - a
time series problem.

May 25, 2019 Slide 13


Heteroskedasticity: Causes
• It may be caused by:
– Model misspecification - omitted variable or improper
functional form.
– Learning behaviors across time as the number of hours of
typing practice increases, the average number of typing
errors as well as their variances decreases.
– Changes in data collection or definitions. companies with
larger profits are generally expected to show greater
variability in their dividend policies than companies with
lower profits. Also, growth oriented companies are likely to
show more variability in their dividend payout ratio than
established companies.
– etc).

May 25, 2019 Slide 14


Outliers
– Outliers or breakdown in model.
• An outlying observation, or outlier, is an observation
that is much different (either very small or very large) in
relation to the observations in the sample. More
precisely, an outlier is an observation from a different
population to that generating the remaining sample
observations.
• Frequently observed in cross sectional data sets where
demographics are involved (population, GNP,
Wrong functional form
Heteroskedasticity: Tests
• Informal Methods
– Plot the data and look for patterns!
– Plot the residuals by the predicted dependent
variable (Resids on the Y-axis)
• Plotting the squared residuals actually makes more
sense, since that is what the assumption refers to!
– Homoskedasticity will be a random scatter
horizontally across the plot.

May 25, 2019 Slide 17


Graphical Representation
Heteroskedasticity: Tests (cont.)
• Park test
– As an exploratory test, log the residuals and
regress them on the logged values of the
suspected independent variable.

ln ui 2  ln  2  B ln X i  vi
 a  B ln X i  vi
– If the B is significant, then Heteroskedasticity may
be a problem.

May 25, 2019 Slide 19


Homoskedasticity: Tests (cont.)
• Goldfeld-Quandt test
– Order the n cases by the X that you think is
correlated with ei2.
– Drop a section of c cases out of the middle
(one-fifth is a reasonable number).
– Run separate regressions on both upper and lower
samples.

May 25, 2019 Slide 20


Heteroskedasticity: Tests (cont.)
• Goldfeld-Quandt test (cont.)
• Do F-test for difference in error variances
F-value has (n - c - 2k)/2 degrees of freedom for each

see1
F n c  2 k n c  2 k 
(
2
,
2
) see2

May 25, 2019 Slide 21


White’s General Heteroscedasticity Test:

• Consider the following regression model



• The White test procedure as follows

• Step-1: By using OLS run and obtain the
residuals .
• Step-2: Then we run the following (auxiliary)
regression
Obtaining Robust errors

• In regression process on the Robust errors


check so that the error term is pressed to
original one . It will make data free from
hetroscadesticity.

May 25, 2019 Slide 23

You might also like