SlideShare a Scribd company logo
8. Heteroskedasticity
We have already seen that homoskedasticity
exists when the error term’s variance, conditional
on all x variables, is constant:
2
)
|
( 

X
u
Var
Homoskedasticity fails if the variance of the error
term varies in the sample (ie: varies with the x
variables)
-We used Homoskedasticity for t tests, F test, and
confidence intervals, even with large samples
8. Heteroskedasticity
8.1 Consequences of Heteroskedasticity for
OLS
8.2 Heteroskedasticity-Robust Inference
after OLS Estimation
8.3 Testing for Heteroskedasticity
8.4 Weighted Least Squares Estimation
8.5 The Linear Probability Model Revisited
8.1 Consequences of Heteroskedasticity
We have already seen that Heteroskedasticity:
1) Does not cause bias or inconsistency (this
depends on MLR. 1 through MLR. 4)
2) Does not affect R2 or adjusted R2 (since these
estimate the POPULATION variances which
are not conditional on X)
Heteroskedasticity does:
1) Make Var(Bjhat) biased, and therefore
invalidate typical OLS standard errors (and
therefore tests)
2) Make OLS no longer BLUE (a better estimator
may exist)
8.2 Heteroskedasticity-Robust Inference
after OLS Estimation
-Because testing hypothesis is a key element of
econometrics, we need to obtain accurate
standard errors in the presence of
heteroskedasticity
-in the last few decades, econometricians have
learned how to adjust standard errors when
HETEROSKEDASTICITY OF UNKNOWN FORM
exists
-these heteroskedasticity-robust procedures are
valid (in large samples) regardless of eror
variance
8.2 Het Fixing 1
-Given a typical single independent variable
model, heteroskedasticity implies a varying
variance:
2
1
0
)
|
( i
i
i
i
i
i
x
u
Var
u
x
y







-Rewriting the OLS slope estimator, we can
obtain a formula for its variance:
2
2
2
1
2
1
1
)
(
)
ˆ
(
)
(
)
(
ˆ
x
i
i
i
i
i
SST
x
x
Var
x
x
u
x
x













8.2 Het Fixing 1
-Recall that  
 2
)
( x
x
SST i
x
-Also notice that given homoskedasticity,
x
i
i
SST
Var
2
1
2
2
)
ˆ
(






-While we don’t know σi
2, White (1980) showed
that a valid estimator is:
2
2
2
1
ˆ
)
(
)
ˆ
(
ˆ
x
i
i
SST
u
x
x
r
a
V
 


8.2 Het Fixing 1
-Given a multiple independent variable model:
u
x
x
y k
k 



 

 ...
1
1
0
-The valid estimator of Var(Bjhat) becomes:
(8.4)
ˆ
ˆ
)
ˆ
(
ˆ 2
2
2
j
i
j
SSR
u
r
r
a
V
ij



-where rijhat2 is the ith residual of a regression of
xj on all other x variables
-where SSRj is the sum of the squared residuals
from that regression
8.2 Het Fixing 1
-The square root of this estimate of variance is
commonly called the HETEROSKEDASTICITY-
ROBUST STANDARD ERROR, but is also called
the White, Huber, or Eickert standard errors due
to its founders
-there are a variety of slight adjustments to this
standard error, but economists generally simply
use the values reported by their program
-this se adjustment gives us
HETEROSKEDASTICITY-ROBUST T STATISTICS:
error
standard
value)
ed
hypothesiz
(estimate 

t
8.2 Why Bother with Normal Errors?
-One may ask why we bother with normal OLS
errors when heteroskedasticity-robust
standard errors are valid more often:
1) Normal OLS t stats have an exact t
distribution, regardless of sample size
2) Robust t statistics are valid only for large
sample sizes
Note that HETEROSKEDASTICITY-ROBUST F
STATISTICS also exist, often called the
HETEROSKEDASTICITY-ROBUST WALD
STATISTIC and reported by most econ
programs.
8.3 Testing for Heteroskedasticity
-In this chapter we will cover a variety of
modern tests for heteroskedasticity
-It is important to know if heteroskedasticity
exists, as its existence means OLS is no
longer the BEST estimator
-Note that while other tests for
heteroskedasticity exist, the test presented
here are preferred due to their more
DIRECT testing for heteroskedasticity
8.3 Testing for Het
-Consider our typical linear model and a null
hypothesis suggesting homoskedasticity:
2
,
2
1
0
1
1
0
)
...,
,
|
(
:
...










k
k
k
x
x
x
u
Var
H
u
x
x
y
Since we know that Var(u|X)=E(u2|X), we can
rewrite the null hypothesis to read:
2
2
,
2
1
2
0 )
(
)
...,
,
|
(
: 

 u
E
x
x
x
u
E
H k
8.3 Testing for Het
-As we are testing whether u2 is related to any
explanatory variables, we can use the linear
model:
v
x
x
u k
k 



 

 ...
1
1
0
2
-where v is an error term with mean zero given
the x’s
-note that the dependent variable is SQUARED
-this changes our null hypothesis to:
0
...
: 2
1
0 


 k
H 


8.3 Testing for Het
-Since we don’t know the true error of the
regression, but only the residual, our estimation
becomes:
error
x
x
u k
k 



 

 ...
ˆ 1
1
0
2
-Which is valid for large sample distributions
-The R2 from the above regression is used to
construct an F statistic:
(8.15)
)
1
/(
)
1
(
/
2
ˆ
2
ˆ
2
2




k
n
R
k
R
F
u
u
8.3 Testing for Het
-This test F statistic is compared to a critical F*
with k, n-k-1 degrees of freedom
-If the null hypothesis is rejected, there is
evidence to conclude that heteroskedasticity
exists at a given α
-If the null hypothesis is not rejected, there is
insufficient evidence to conclude that
heteroskedasticity exists at a given α
-this is sometimes called the BREUCH-PAGAN
TEST FOR HETEROSKEDASTICITY (BP TEST)
8.3 BP HET TEST
In order to conduct a BP test for het
1) Run a normal OLS regression (y on x’s) and
obtain the square of the residuals, uhat2
2) Run a regression of uhat2 on all independent
variables and save the R2
3) Obtain a test F statistic and compare it to the
critical F*
4) If F>F*, reject the null hypothesis of
homoskedasticity and start correcting for
heteroskedasticity
8.3 BP HET TEST
If we suspect that our model’s heteroskedasticity
depends on only certain x variables,
Only regress uhat2 on those variables
-Keep in mind that the K in the R2 formula and in
the degrees of freedom comes from the
number of independent variables in the uhat2
regression
An alternate test for het is the white test:
8.3 White Test for Het
-Given the statistical modifications covered in
chapter 5, White (1980) proposed another test
for heteroskedasticity
-With 3 independent variables, White proposed a
linear regression with 9 regressors:
error
x
x
x
x
x
x
x
x
x
x
x
x
u











3
2
9
3
1
8
2
1
7
2
3
6
2
2
5
2
1
4
3
3
2
2
1
1
0
2
ˆ










-The null hypothesis (homoskedasticity) now sets
all δ (except the intercept) equal to zero
8.3 White Test for Het
-Unfortunately this test involves MANY regressors
(27 regressors for 6 x variables) and as such
may have degrees of freedom issues
-one special case of the White test is to estimate
the regression:
error
y
y
u 


 2
2
1
0
2
ˆ
ˆ
ˆ 


-since this preserves the “squared” concept of
the White test and is particularly useful when
het is suspected to be connected to the level of
the expected value E(y|X)
-this test has a F distribution w/2,n-3 df
8.3 Special White HET TEST
In order to conduct a special White test for het
1) Run a normal OLS regression (y on x’s) and
obtain the square of the residuals, uhat2 and
the predicted values, yhat
2) Run the regression of uhat2 on both yhat and
yhat2 (including an intercept). Record the R2
values
3) Using these R2 values, compute a test F
statistic as in the BP test
4) If F>F*, reject the null hypothesis
(homoskedasticity)
8.3 Heteroskedasticity Note
-Our decision to REJECT the null hypothesis and
suspect heteroskedasticity is only valid if MLR.4
is valid
-if MLR.4 is violated (ie: bad funcitonal form or
omitted variables), one can reject the null
hypothesis even if het doesn’t actually exist
-Therefore always chose functional form and all
variables before testing for heteroskedasticity

More Related Content

Similar to Econ 399 Chapter8a.ppt

Chi square test
Chi square testChi square test
Chi square test
Patel Parth
 
Multivariate reg analysis
Multivariate reg analysisMultivariate reg analysis
Multivariate reg analysis
Irfan Hussain
 
Priliminary Research on Multi-Dimensional Panel Data Modeling
Priliminary Research on Multi-Dimensional Panel Data Modeling Priliminary Research on Multi-Dimensional Panel Data Modeling
Priliminary Research on Multi-Dimensional Panel Data Modeling
Parthasarathi E
 
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
andifebby2
 
Prob and statistics models for outlier detection
Prob and statistics models for outlier detectionProb and statistics models for outlier detection
Prob and statistics models for outlier detection
Trilochan Panigrahi
 
Looking for cooperation on working paper - Expenditure model
Looking for cooperation on working paper - Expenditure modelLooking for cooperation on working paper - Expenditure model
Looking for cooperation on working paper - Expenditure model
Miss. Antónia FICOVÁ, Engineer. (Not yet Dr.)
 
Chapter8
Chapter8Chapter8
Chapter8
Vu Vo
 
Regression Analysis
Regression AnalysisRegression Analysis
Regression Analysis
Michael770443
 
2013poster
2013poster2013poster
2013poster
Ian Finley
 
A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4
Mintu246
 
A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4
Mintu246
 
Regression vs Neural Net
Regression vs Neural NetRegression vs Neural Net
Regression vs Neural Net
Ratul Alahy
 
ImperialMathFinance: Finance and Stochastics Seminar
ImperialMathFinance: Finance and Stochastics SeminarImperialMathFinance: Finance and Stochastics Seminar
ImperialMathFinance: Finance and Stochastics Seminar
elviszhang
 
I stata
I stataI stata
I stata
Hien Hai Bui
 
Diagnostic methods for Building the regression model
Diagnostic methods for Building the regression modelDiagnostic methods for Building the regression model
Diagnostic methods for Building the regression model
Mehdi Shayegani
 
Practice test ch 10 correlation reg ch 11 gof ch12 anova
Practice test ch 10 correlation reg ch 11 gof ch12 anovaPractice test ch 10 correlation reg ch 11 gof ch12 anova
Practice test ch 10 correlation reg ch 11 gof ch12 anova
Long Beach City College
 
regression-130929093340-phpapp02 (1).pdf
regression-130929093340-phpapp02 (1).pdfregression-130929093340-phpapp02 (1).pdf
regression-130929093340-phpapp02 (1).pdf
MuhammadAftab89
 
Advanced Econometrics L5-6.pptx
Advanced Econometrics L5-6.pptxAdvanced Econometrics L5-6.pptx
Advanced Econometrics L5-6.pptx
akashayosha
 
Reg
RegReg
isi
isiisi
isi
ahfdv
 

Similar to Econ 399 Chapter8a.ppt (20)

Chi square test
Chi square testChi square test
Chi square test
 
Multivariate reg analysis
Multivariate reg analysisMultivariate reg analysis
Multivariate reg analysis
 
Priliminary Research on Multi-Dimensional Panel Data Modeling
Priliminary Research on Multi-Dimensional Panel Data Modeling Priliminary Research on Multi-Dimensional Panel Data Modeling
Priliminary Research on Multi-Dimensional Panel Data Modeling
 
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
1648796607723_Material-8---Concept-on-Estimation-Variance.pdf
 
Prob and statistics models for outlier detection
Prob and statistics models for outlier detectionProb and statistics models for outlier detection
Prob and statistics models for outlier detection
 
Looking for cooperation on working paper - Expenditure model
Looking for cooperation on working paper - Expenditure modelLooking for cooperation on working paper - Expenditure model
Looking for cooperation on working paper - Expenditure model
 
Chapter8
Chapter8Chapter8
Chapter8
 
Regression Analysis
Regression AnalysisRegression Analysis
Regression Analysis
 
2013poster
2013poster2013poster
2013poster
 
A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4
 
A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4A comparative analysis of predictve data mining techniques4
A comparative analysis of predictve data mining techniques4
 
Regression vs Neural Net
Regression vs Neural NetRegression vs Neural Net
Regression vs Neural Net
 
ImperialMathFinance: Finance and Stochastics Seminar
ImperialMathFinance: Finance and Stochastics SeminarImperialMathFinance: Finance and Stochastics Seminar
ImperialMathFinance: Finance and Stochastics Seminar
 
I stata
I stataI stata
I stata
 
Diagnostic methods for Building the regression model
Diagnostic methods for Building the regression modelDiagnostic methods for Building the regression model
Diagnostic methods for Building the regression model
 
Practice test ch 10 correlation reg ch 11 gof ch12 anova
Practice test ch 10 correlation reg ch 11 gof ch12 anovaPractice test ch 10 correlation reg ch 11 gof ch12 anova
Practice test ch 10 correlation reg ch 11 gof ch12 anova
 
regression-130929093340-phpapp02 (1).pdf
regression-130929093340-phpapp02 (1).pdfregression-130929093340-phpapp02 (1).pdf
regression-130929093340-phpapp02 (1).pdf
 
Advanced Econometrics L5-6.pptx
Advanced Econometrics L5-6.pptxAdvanced Econometrics L5-6.pptx
Advanced Econometrics L5-6.pptx
 
Reg
RegReg
Reg
 
isi
isiisi
isi
 

Recently uploaded

Earn a passive income with prosocial investing
Earn a passive income with prosocial investingEarn a passive income with prosocial investing
Earn a passive income with prosocial investing
Colin R. Turner
 
Bridging the gap: Online job postings, survey data and the assessment of job ...
Bridging the gap: Online job postings, survey data and the assessment of job ...Bridging the gap: Online job postings, survey data and the assessment of job ...
Bridging the gap: Online job postings, survey data and the assessment of job ...
Labour Market Information Council | Conseil de l’information sur le marché du travail
 
An Overview of the Prosocial dHEDGE Vault works
An Overview of the Prosocial dHEDGE Vault worksAn Overview of the Prosocial dHEDGE Vault works
An Overview of the Prosocial dHEDGE Vault works
Colin R. Turner
 
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
nimaruinazawa258
 
Instant Issue Debit Cards
Instant Issue Debit CardsInstant Issue Debit Cards
Instant Issue Debit Cards
egoetzinger
 
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
5spllj1l
 
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
qntjwn68
 
falcon-invoice-discounting-a-strategic-approach-to-optimize-investments
falcon-invoice-discounting-a-strategic-approach-to-optimize-investmentsfalcon-invoice-discounting-a-strategic-approach-to-optimize-investments
falcon-invoice-discounting-a-strategic-approach-to-optimize-investments
Falcon Invoice Discounting
 
1.2 Business Ideas Business Ideas Busine
1.2 Business Ideas Business Ideas Busine1.2 Business Ideas Business Ideas Busine
1.2 Business Ideas Business Ideas Busine
Lawrence101
 
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
mayaclinic18
 
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
obyzuk
 
Economic Risk Factor Update: June 2024 [SlideShare]
Economic Risk Factor Update: June 2024 [SlideShare]Economic Risk Factor Update: June 2024 [SlideShare]
Economic Risk Factor Update: June 2024 [SlideShare]
Commonwealth
 
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
otogas
 
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
rlo9fxi
 
一比一原版(IC毕业证)帝国理工大学毕业证如何办理
一比一原版(IC毕业证)帝国理工大学毕业证如何办理一比一原版(IC毕业证)帝国理工大学毕业证如何办理
一比一原版(IC毕业证)帝国理工大学毕业证如何办理
conose1
 
Applying the Global Internal Audit Standards_AIS.pdf
Applying the Global Internal Audit Standards_AIS.pdfApplying the Global Internal Audit Standards_AIS.pdf
Applying the Global Internal Audit Standards_AIS.pdf
alexiusbrian1
 
1. Elemental Economics - Introduction to mining.pdf
1. Elemental Economics - Introduction to mining.pdf1. Elemental Economics - Introduction to mining.pdf
1. Elemental Economics - Introduction to mining.pdf
Neal Brewster
 
How Non-Banking Financial Companies Empower Startups With Venture Debt Financing
How Non-Banking Financial Companies Empower Startups With Venture Debt FinancingHow Non-Banking Financial Companies Empower Startups With Venture Debt Financing
How Non-Banking Financial Companies Empower Startups With Venture Debt Financing
Vighnesh Shashtri
 
Who Is the Largest Producer of Soybean in India Now.pdf
Who Is the Largest Producer of Soybean in India Now.pdfWho Is the Largest Producer of Soybean in India Now.pdf
Who Is the Largest Producer of Soybean in India Now.pdf
Price Vision
 
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
bbeucd
 

Recently uploaded (20)

Earn a passive income with prosocial investing
Earn a passive income with prosocial investingEarn a passive income with prosocial investing
Earn a passive income with prosocial investing
 
Bridging the gap: Online job postings, survey data and the assessment of job ...
Bridging the gap: Online job postings, survey data and the assessment of job ...Bridging the gap: Online job postings, survey data and the assessment of job ...
Bridging the gap: Online job postings, survey data and the assessment of job ...
 
An Overview of the Prosocial dHEDGE Vault works
An Overview of the Prosocial dHEDGE Vault worksAn Overview of the Prosocial dHEDGE Vault works
An Overview of the Prosocial dHEDGE Vault works
 
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
Tdasx: In-Depth Analysis of Cryptocurrency Giveaway Scams and Security Strate...
 
Instant Issue Debit Cards
Instant Issue Debit CardsInstant Issue Debit Cards
Instant Issue Debit Cards
 
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
在线办理(GU毕业证书)美国贡萨加大学毕业证学历证书一模一样
 
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
1:1制作加拿大麦吉尔大学毕业证硕士学历证书原版一模一样
 
falcon-invoice-discounting-a-strategic-approach-to-optimize-investments
falcon-invoice-discounting-a-strategic-approach-to-optimize-investmentsfalcon-invoice-discounting-a-strategic-approach-to-optimize-investments
falcon-invoice-discounting-a-strategic-approach-to-optimize-investments
 
1.2 Business Ideas Business Ideas Busine
1.2 Business Ideas Business Ideas Busine1.2 Business Ideas Business Ideas Busine
1.2 Business Ideas Business Ideas Busine
 
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
^%$Zone1:+971)581248768’][* Legit & Safe #Abortion #Pills #For #Sale In #Duba...
 
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
一比一原版(GWU,GW毕业证)加利福尼亚大学|尔湾分校毕业证如何办理
 
Economic Risk Factor Update: June 2024 [SlideShare]
Economic Risk Factor Update: June 2024 [SlideShare]Economic Risk Factor Update: June 2024 [SlideShare]
Economic Risk Factor Update: June 2024 [SlideShare]
 
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
一比一原版(UCL毕业证)伦敦大学|学院毕业证如何办理
 
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
快速制作美国迈阿密大学牛津分校毕业证文凭证书英文原版一模一样
 
一比一原版(IC毕业证)帝国理工大学毕业证如何办理
一比一原版(IC毕业证)帝国理工大学毕业证如何办理一比一原版(IC毕业证)帝国理工大学毕业证如何办理
一比一原版(IC毕业证)帝国理工大学毕业证如何办理
 
Applying the Global Internal Audit Standards_AIS.pdf
Applying the Global Internal Audit Standards_AIS.pdfApplying the Global Internal Audit Standards_AIS.pdf
Applying the Global Internal Audit Standards_AIS.pdf
 
1. Elemental Economics - Introduction to mining.pdf
1. Elemental Economics - Introduction to mining.pdf1. Elemental Economics - Introduction to mining.pdf
1. Elemental Economics - Introduction to mining.pdf
 
How Non-Banking Financial Companies Empower Startups With Venture Debt Financing
How Non-Banking Financial Companies Empower Startups With Venture Debt FinancingHow Non-Banking Financial Companies Empower Startups With Venture Debt Financing
How Non-Banking Financial Companies Empower Startups With Venture Debt Financing
 
Who Is the Largest Producer of Soybean in India Now.pdf
Who Is the Largest Producer of Soybean in India Now.pdfWho Is the Largest Producer of Soybean in India Now.pdf
Who Is the Largest Producer of Soybean in India Now.pdf
 
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
一比一原版(UCSB毕业证)圣芭芭拉分校毕业证如何办理
 

Econ 399 Chapter8a.ppt

  • 1. 8. Heteroskedasticity We have already seen that homoskedasticity exists when the error term’s variance, conditional on all x variables, is constant: 2 ) | (   X u Var Homoskedasticity fails if the variance of the error term varies in the sample (ie: varies with the x variables) -We used Homoskedasticity for t tests, F test, and confidence intervals, even with large samples
  • 2. 8. Heteroskedasticity 8.1 Consequences of Heteroskedasticity for OLS 8.2 Heteroskedasticity-Robust Inference after OLS Estimation 8.3 Testing for Heteroskedasticity 8.4 Weighted Least Squares Estimation 8.5 The Linear Probability Model Revisited
  • 3. 8.1 Consequences of Heteroskedasticity We have already seen that Heteroskedasticity: 1) Does not cause bias or inconsistency (this depends on MLR. 1 through MLR. 4) 2) Does not affect R2 or adjusted R2 (since these estimate the POPULATION variances which are not conditional on X) Heteroskedasticity does: 1) Make Var(Bjhat) biased, and therefore invalidate typical OLS standard errors (and therefore tests) 2) Make OLS no longer BLUE (a better estimator may exist)
  • 4. 8.2 Heteroskedasticity-Robust Inference after OLS Estimation -Because testing hypothesis is a key element of econometrics, we need to obtain accurate standard errors in the presence of heteroskedasticity -in the last few decades, econometricians have learned how to adjust standard errors when HETEROSKEDASTICITY OF UNKNOWN FORM exists -these heteroskedasticity-robust procedures are valid (in large samples) regardless of eror variance
  • 5. 8.2 Het Fixing 1 -Given a typical single independent variable model, heteroskedasticity implies a varying variance: 2 1 0 ) | ( i i i i i i x u Var u x y        -Rewriting the OLS slope estimator, we can obtain a formula for its variance: 2 2 2 1 2 1 1 ) ( ) ˆ ( ) ( ) ( ˆ x i i i i i SST x x Var x x u x x             
  • 6. 8.2 Het Fixing 1 -Recall that    2 ) ( x x SST i x -Also notice that given homoskedasticity, x i i SST Var 2 1 2 2 ) ˆ (       -While we don’t know σi 2, White (1980) showed that a valid estimator is: 2 2 2 1 ˆ ) ( ) ˆ ( ˆ x i i SST u x x r a V    
  • 7. 8.2 Het Fixing 1 -Given a multiple independent variable model: u x x y k k         ... 1 1 0 -The valid estimator of Var(Bjhat) becomes: (8.4) ˆ ˆ ) ˆ ( ˆ 2 2 2 j i j SSR u r r a V ij    -where rijhat2 is the ith residual of a regression of xj on all other x variables -where SSRj is the sum of the squared residuals from that regression
  • 8. 8.2 Het Fixing 1 -The square root of this estimate of variance is commonly called the HETEROSKEDASTICITY- ROBUST STANDARD ERROR, but is also called the White, Huber, or Eickert standard errors due to its founders -there are a variety of slight adjustments to this standard error, but economists generally simply use the values reported by their program -this se adjustment gives us HETEROSKEDASTICITY-ROBUST T STATISTICS: error standard value) ed hypothesiz (estimate   t
  • 9. 8.2 Why Bother with Normal Errors? -One may ask why we bother with normal OLS errors when heteroskedasticity-robust standard errors are valid more often: 1) Normal OLS t stats have an exact t distribution, regardless of sample size 2) Robust t statistics are valid only for large sample sizes Note that HETEROSKEDASTICITY-ROBUST F STATISTICS also exist, often called the HETEROSKEDASTICITY-ROBUST WALD STATISTIC and reported by most econ programs.
  • 10. 8.3 Testing for Heteroskedasticity -In this chapter we will cover a variety of modern tests for heteroskedasticity -It is important to know if heteroskedasticity exists, as its existence means OLS is no longer the BEST estimator -Note that while other tests for heteroskedasticity exist, the test presented here are preferred due to their more DIRECT testing for heteroskedasticity
  • 11. 8.3 Testing for Het -Consider our typical linear model and a null hypothesis suggesting homoskedasticity: 2 , 2 1 0 1 1 0 ) ..., , | ( : ...           k k k x x x u Var H u x x y Since we know that Var(u|X)=E(u2|X), we can rewrite the null hypothesis to read: 2 2 , 2 1 2 0 ) ( ) ..., , | ( :    u E x x x u E H k
  • 12. 8.3 Testing for Het -As we are testing whether u2 is related to any explanatory variables, we can use the linear model: v x x u k k         ... 1 1 0 2 -where v is an error term with mean zero given the x’s -note that the dependent variable is SQUARED -this changes our null hypothesis to: 0 ... : 2 1 0     k H   
  • 13. 8.3 Testing for Het -Since we don’t know the true error of the regression, but only the residual, our estimation becomes: error x x u k k         ... ˆ 1 1 0 2 -Which is valid for large sample distributions -The R2 from the above regression is used to construct an F statistic: (8.15) ) 1 /( ) 1 ( / 2 ˆ 2 ˆ 2 2     k n R k R F u u
  • 14. 8.3 Testing for Het -This test F statistic is compared to a critical F* with k, n-k-1 degrees of freedom -If the null hypothesis is rejected, there is evidence to conclude that heteroskedasticity exists at a given α -If the null hypothesis is not rejected, there is insufficient evidence to conclude that heteroskedasticity exists at a given α -this is sometimes called the BREUCH-PAGAN TEST FOR HETEROSKEDASTICITY (BP TEST)
  • 15. 8.3 BP HET TEST In order to conduct a BP test for het 1) Run a normal OLS regression (y on x’s) and obtain the square of the residuals, uhat2 2) Run a regression of uhat2 on all independent variables and save the R2 3) Obtain a test F statistic and compare it to the critical F* 4) If F>F*, reject the null hypothesis of homoskedasticity and start correcting for heteroskedasticity
  • 16. 8.3 BP HET TEST If we suspect that our model’s heteroskedasticity depends on only certain x variables, Only regress uhat2 on those variables -Keep in mind that the K in the R2 formula and in the degrees of freedom comes from the number of independent variables in the uhat2 regression An alternate test for het is the white test:
  • 17. 8.3 White Test for Het -Given the statistical modifications covered in chapter 5, White (1980) proposed another test for heteroskedasticity -With 3 independent variables, White proposed a linear regression with 9 regressors: error x x x x x x x x x x x x u            3 2 9 3 1 8 2 1 7 2 3 6 2 2 5 2 1 4 3 3 2 2 1 1 0 2 ˆ           -The null hypothesis (homoskedasticity) now sets all δ (except the intercept) equal to zero
  • 18. 8.3 White Test for Het -Unfortunately this test involves MANY regressors (27 regressors for 6 x variables) and as such may have degrees of freedom issues -one special case of the White test is to estimate the regression: error y y u     2 2 1 0 2 ˆ ˆ ˆ    -since this preserves the “squared” concept of the White test and is particularly useful when het is suspected to be connected to the level of the expected value E(y|X) -this test has a F distribution w/2,n-3 df
  • 19. 8.3 Special White HET TEST In order to conduct a special White test for het 1) Run a normal OLS regression (y on x’s) and obtain the square of the residuals, uhat2 and the predicted values, yhat 2) Run the regression of uhat2 on both yhat and yhat2 (including an intercept). Record the R2 values 3) Using these R2 values, compute a test F statistic as in the BP test 4) If F>F*, reject the null hypothesis (homoskedasticity)
  • 20. 8.3 Heteroskedasticity Note -Our decision to REJECT the null hypothesis and suspect heteroskedasticity is only valid if MLR.4 is valid -if MLR.4 is violated (ie: bad funcitonal form or omitted variables), one can reject the null hypothesis even if het doesn’t actually exist -Therefore always chose functional form and all variables before testing for heteroskedasticity