9 views

Uploaded by Mark Ebrahim

What is Binomial data?

save

- Customer Base Analysis Notes
- Asian Economic Journal1
- tmp29D4.tmp
- A Survey on Gas Leakage Source Detection and Boundary Tracking With Wireless Sensor Networks
- Hydrological Model Sensitivity to Parameter and Radar Rainfall Estimation Uncertainty
- Continent-wide Survey Reveals Massive Decline in African Savannah Elephants
- Econometrics Assignment HW4
- A Review of Barker’s Activity Coefficient Method and VLE Data Reduction
- Presentation for data mining
- DSmT Based RX Detector for Hyperspectral Imagery
- Romero 2014sudahprint
- Diebolt, Guiraud (2005) a Note on Long Memory Time Series
- Mix Operational Risk
- Curs 1 SSL - Introduction
- Lecture e10
- myello
- ch05-solns-all_skuce_2e
- great project stats
- MI Paper 2 Solns.pdf
- technote2_betabinomial
- Breezer Report
- 173232298-A-Guide-to-Modern-Econometrics-by-Verbeek_171-180.pdf
- 1311.2878.pdf
- MJC JC 2 H2 Maths 2011 Mid Year Exam Solutions Paper 2
- Hayward Green Vg
- final project
- 6 sigma-9
- Abstrak 25 Juli
- CH16
- ch7 and ch8
- Ethiopia CRGE Strategy Final
- THE LEAST DEVELOPED COUNTRIES REPORT 2012
- Growth and Transformation Plan-Ethiopian
- Boys and girls in Ethiopia
- Improving drought response in pastoral areas of Ethiopia
- asc-07Revisiting resistance in Italian-occupied Ethiopia: The Patriots’ Movement (1936-1941) and the redefinition of post-war Ethiopia 5287668-243-01
- Ethiopia FINAL August 2012_LM
- Ethiopia's profile
- ANOVA
- ANOVA_how_to_do
- Poverty Reduction Strategy Paper of the gov't of Ethiopia
- The Criminal Code of The Federal Democratic Republic of Ethiopia (2004)
- CoConstitution of The Federal Democratic Republic of Ethiopianstitution Ethiopia
- Ethiopia’s Grand rEnaissancE
- Black Gold for Blue Gold? Sudan’s Oil, Ethiopia’s Water and Regional Integration
- Oven_Recipe_Cook_Book
- Medium Term
- Chapter 05 Inputs & Costs.pptx
- Introduction to Statistics
- Driving in Finland
- Ethiopia English
- Demographic and Health Survey 2011 - Ethiopia
- Introduction to Database Managment
- Applied Statistics for Bioinformatics using_R
- Introduction to Biometrics
- Vuokrasopimuksen esisopimus
- CancerEpi.pdf
- Mastawesha=3-1.pdf
- Final Nutrition Cluster 2012 Evaluation Report 2013-02-20
- MPH Curriculum Example

You are on page 1of 4

3-3

•A linear model approach for binomial data:

y

x

/n

x

=X| +c, c ~ N (0, o

2

I )

Q: when is the approach appropriate?

The pmf shape of the joint distribution of y

x

/n

x

is similar to the pdf shape of N (X|, o

2

I )

Some problems with this approach

Predicted probability may >1 or <0

Normal approximation might be too

much a stretch when n

i

’s are not

large or p

x

~1/0

Variance of Binomial is not constant

Some of these problems could be

corrected by using transformation

and weighting

p. 3-4

•Recall: linear model

model description 1: Y =X| +c, c ~ N (0, o

2

I )

model description 2: Y ~ N (X|, o

2

I )

Y ~ N (µ

x

, E

x

)

µ

x

=q

x

, E

x

=o

2

I

Q: which description can be generalized to binomial data?

•3 components in a generalized linear model (binomial example)

y

x

~B(n

x

, p

x

)

link functiong: g monotone and q

x

=g(p

x

) [for binimial,

g: (0, 1)÷(÷·, ·) ]

•Common choices of link function for binomial data

Logit: q

x

=log(p

x

/(1÷p

x

))

Probit: q

x

=u

÷1

(p

x

), where u is the cdf of Normal

Xβ =

p

i=1

β

i

· h

i

(X

1

, . . . , X

m

) ≡ η

x

Xβ =

p

i=1

β

i

· h

i

(X

1

, . . . , X

m

) ≡ η

x

, −∞< η

x

< ∞

Generalized Linear Model for Binomial Data

NTHU STAT 5230, 2011 Lecture Notes

made by Shao-Wei Cheng (NTHU)

p. 3-5

Complementary log-log: q

x

=log(÷log(1÷p

x

))

Logit is close to the complementary log-log when p

x

is small

Logit is close to probit when 0.1 <p

x

<0.9

(exercise: plot & check the difference between the 3 functions)

•Log-likelihood of the GLM (use logit link as an example):

•Estimation of |

Recall: in linear model, the LS estimator of | is also the MLE

For GLM, the concept of LS does not work any more ¬still

can adopt the method of MLE (maximize l(|) as a function of |)

Usually no explicit formula for MLE of |

An algorithm to perform the maximization will be discussed in

future lecture

p. 3-6

I nference

•RSS and deviance

Recall: in linear model, RSS play a critical role in inference

Q: In GLM, what is the concept similar to RSS?

consider two models

a larger model L with l parameters and likelihood L

L

a smaller model S with s parameters and likelihood L

S

S is nested in L (S c L)

To test H

0

: S v.s. H

1

:L\S, likelihood methods suggests the

likelihood ratio statistics:

2 log(L

L

/L

S

)

suppose that L is a saturated larger model, the test statistic

becomes:

where is the fitted values from the smaller model

D is called deviance, which plays a role similar to RSS

2 log(L

L

(

ˆ

β

L

)/L

S

(

ˆ

β

S

))

NTHU STAT 5230, 2011 Lecture Notes

made by Shao-Wei Cheng (NTHU)

p. 3-7

Since the saturated model fits as well as any model can fit, the

deviance D measures how close the (smaller) model comes to

perfection.

Deviance can be treated as a measure of goodness of fit

•Suppose that y

i

is truly binomial and that the n

i

are relatively large

, if the (smaller) model is correct ¬can use the

deviance to test whether the model is an adequate fit

The chi-square distribution is only an approximation that

becomes more accurate as the n

i

increase [often suggest n

i

> 5]

•Use deviance to compare two models S and L, S nested inL

Larger model L: deviance D

L

and df

L

(=k÷l)

Smaller model S: deviance D

S

and df

S

(=k÷s)

To test H

0

: S v.s. H

1

:L\S, the test statistics is

D

S

÷ D

L

which is asymptotically distributed as

•In terms of the accuracy of dist. approx., test >goodness of fit

p. 3-8

•(Wald’s test) alternative test for H

0

: |

i

=0

Can be generalized to H

0

: |

i

=c or H

0

: | =c

Asymptotic null distribution: N(0, 1)

in contrast to normal linear model, these two statistics

(deviance-based and Wald’s tests) are not identical

Hauck-Donner effect (see Hauck and Donner, 1977): for

sparse data(i.e., many n

i

’s =1 or small), the standard errors

can be overestimatedand so the z-value is too small and the

significance of an effect could be missed

¬therefore, the deviance-based test is preferred

test statistics: z-value

NTHU STAT 5230, 2011 Lecture Notes

made by Shao-Wei Cheng (NTHU)

p. 3-9

•100(1÷o)% confidence interval

Relationship between confidence interval and test

Approach 1: (from Wald’s test)

Approach 2: (profile likelihood-based method)

other |

j

’s, j=i, set to the maximizing values

(recall: the computation of the C.I. for ì in Box-Cox method)

the profile likelihood method is generally preferable for the

same Hauck-Donner reason

Similar method can be generalized to construct confidence

region of several parameters

NTHU STAT 5230, 2011 Lecture Notes

made by Shao-Wei Cheng (NTHU)

- Customer Base Analysis NotesUploaded byRoss Kendall
- Asian Economic Journal1Uploaded bysekar_smr
- tmp29D4.tmpUploaded byFrontiers
- A Survey on Gas Leakage Source Detection and Boundary Tracking With Wireless Sensor NetworksUploaded byBrenner Biasi Silva
- Hydrological Model Sensitivity to Parameter and Radar Rainfall Estimation UncertaintyUploaded bySudharsananPRS
- Continent-wide Survey Reveals Massive Decline in African Savannah ElephantsUploaded byikutmilis
- Econometrics Assignment HW4Uploaded byNikhil Sharma
- A Review of Barker’s Activity Coefficient Method and VLE Data ReductionUploaded bymurdanetap957
- Presentation for data miningUploaded byNikola
- DSmT Based RX Detector for Hyperspectral ImageryUploaded byscience2010
- Romero 2014sudahprintUploaded byputri nadia
- Diebolt, Guiraud (2005) a Note on Long Memory Time SeriesUploaded byJoão Victor Machado
- Mix Operational RiskUploaded byMeryem Khiari Mohamed Akram
- Curs 1 SSL - IntroductionUploaded byMihai Ilie
- Lecture e10Uploaded byJung Yoon Song
- myelloUploaded byForrest Shaka Nahasapeemapetilon
- ch05-solns-all_skuce_2eUploaded bygainesboro
- great project statsUploaded byapi-275663595
- MI Paper 2 Solns.pdfUploaded byNg Shu Qing
- technote2_betabinomialUploaded byxmanian9
- Breezer ReportUploaded byVINCENYYYYYYY
- 173232298-A-Guide-to-Modern-Econometrics-by-Verbeek_171-180.pdfUploaded byAnonymous T2LhplU
- 1311.2878.pdfUploaded byMuguen8
- MJC JC 2 H2 Maths 2011 Mid Year Exam Solutions Paper 2Uploaded byjimmytanlimlong
- Hayward Green VgUploaded bysumasuthan
- final projectUploaded byapi-240732385
- 6 sigma-9Uploaded bydheenasweet
- Abstrak 25 JuliUploaded byAnyssAdilah
- CH16Uploaded bybihosambi
- ch7 and ch8Uploaded byMachiavelli Herrera

- Ethiopia CRGE Strategy FinalUploaded byMark Ebrahim
- THE LEAST DEVELOPED COUNTRIES REPORT 2012Uploaded byMark Ebrahim
- Growth and Transformation Plan-EthiopianUploaded byMark Ebrahim
- Boys and girls in EthiopiaUploaded byMark Ebrahim
- Improving drought response in pastoral areas of EthiopiaUploaded byMark Ebrahim
- asc-07Revisiting resistance in Italian-occupied Ethiopia: The Patriots’ Movement (1936-1941) and the redefinition of post-war Ethiopia 5287668-243-01Uploaded byMark Ebrahim
- Ethiopia FINAL August 2012_LMUploaded byMark Ebrahim
- Ethiopia's profileUploaded byMark Ebrahim
- ANOVAUploaded byMark Ebrahim
- ANOVA_how_to_doUploaded byMark Ebrahim
- Poverty Reduction Strategy Paper of the gov't of EthiopiaUploaded byMark Ebrahim
- The Criminal Code of The Federal Democratic Republic of Ethiopia (2004)Uploaded byMark Ebrahim
- CoConstitution of The Federal Democratic Republic of Ethiopianstitution EthiopiaUploaded byMark Ebrahim
- Ethiopia’s Grand rEnaissancEUploaded byMark Ebrahim
- Black Gold for Blue Gold? Sudan’s Oil, Ethiopia’s Water and Regional IntegrationUploaded byMark Ebrahim
- Oven_Recipe_Cook_BookUploaded byMark Ebrahim
- Medium TermUploaded byMark Ebrahim
- Chapter 05 Inputs & Costs.pptxUploaded byMark Ebrahim
- Introduction to StatisticsUploaded byMark Ebrahim
- Driving in FinlandUploaded byMark Ebrahim
- Ethiopia EnglishUploaded byMark Ebrahim
- Demographic and Health Survey 2011 - EthiopiaUploaded byMark Ebrahim
- Introduction to Database ManagmentUploaded byMark Ebrahim
- Applied Statistics for Bioinformatics using_RUploaded byMark Ebrahim
- Introduction to BiometricsUploaded byMark Ebrahim
- Vuokrasopimuksen esisopimusUploaded byMark Ebrahim
- CancerEpi.pdfUploaded byMark Ebrahim
- Mastawesha=3-1.pdfUploaded byMark Ebrahim
- Final Nutrition Cluster 2012 Evaluation Report 2013-02-20Uploaded byMark Ebrahim
- MPH Curriculum ExampleUploaded byMark Ebrahim