11 views

Uploaded by missinu

Poe 4 Formulas

- Application-of-Artificial-Intelligence-Approach-in-Modeling-_2016_Procedia-T.pdf
- Solutions Manual Ch13 - 2012
- DeVry MATH 533 Final Exam 100% Correct Answer
- Letter to Shareholders
- Week 2 Individual y Grupal
- Standard Costing
- 143143458 Brand Loyalty
- Chapter 16
- STATISTICS IN EDUCATION.pdf
- Literacy Practices and Schooling
- 2 - Bowman Helfat - Does Corporate Strategy Matter (1)
- Manova x2 x19 x20
- Stats Ch8 Formulas
- jurnal honey comb.pdf
- qtec-35-4-335
- The Evaluation Relationship Between the Factors Affecting the Security of Information Systems and the Financial Reporting
- initial report
- quality manajemen
- research paper samstipic
- 180

You are on page 1of 3

n

i=1

x

i

= x

1

x

2

x

n

n

i=1

a = na

n

i=1

ax

i

= a

n

i=1

x

i

n

i=1

(x

i

y

i

) =

n

i=1

x

i

n

i=1

y

i

n

i=1

(ax

i

by

i

) = a

n

i=1

x

i

b

n

i=1

y

i

n

i=1

(a bx

i

) = na b

n

i=1

x

i

x =

n

i=1

x

i

n

=

x

1

x

2

x

n

n

n

i=1

(x

i

x) = 0

2

i=1

3

j=1

f (x

i

; y

j

) =

2

i=1

f (x

i

; y

1

) f (x

i

; y

2

) f (x

i

; y

3

) [ [

= f (x

1

; y

1

) f (x

1

; y

2

) f (x

1

; y

3

)

f (x

2

; y

1

) f (x

2

; y

2

) f (x

2

; y

3

)

Expected Values & Variances

E(X) = x

1

f (x

1

) x

2

f (x

2

) x

n

f (x

n

)

=

n

i=1

x

i

f (x

i

) =

x

x f (x)

E g(X) [ [ =

x

g(x) f (x)

E g

1

(X) g

2

(X) [ [ =

x

g

1

(x) g

2

(x) [ [ f (x)

=

x

g

1

(x) f (x)

x

g

2

(x) f (x)

= E g

1

(X) [ [ E g

2

(X) [ [

E(c) = c

E(cX) = cE(X)

E(a cX) = a cE(X)

var(X) = s

2

= E[X E(X)]

2

= E(X

2

) [E(X)]

2

var(a cX) = E[(a cX) E(a cX)]

2

= c

2

var(X)

Marginal and Conditional Distributions

f (x) =

y

f (x; y) for each value X can take

f (y) =

x

f (x; y) for each value Y can take

f (x[y) = P X = x[Y = y [ [ =

f (x; y)

f (y)

If X and Y are independent random variables, then

f (x,y) = f (x)f ( y) for each and every pair of values

x and y. The converse is also true.

If X and Y are independent random variables, then the

conditional probability density function of X given that

Y = y is f (x[y) =

f (x; y)

f (y)

=

f (x) f (y)

f (y)

= f (x)

for each and everypair of values x and y. The converse is

also true.

Expectations, Variances & Covariances

cov(X; Y) = E[(XE[X[)(YE[Y[)[

=

x

y

x E(X) [ [ y E(Y) [ [ f (x; y)

r =

cov(X;Y)

var(X)var(Y)

_

E(c

1

X c

2

Y) = c

1

E(X) c

2

E(Y)

E(X Y) = E(X) E(Y)

var(aXbYcZ) =a

2

var(X) b

2

var(Y) c

2

var(Z)

2abcov(X,Y) 2accov(X,Z) 2bccov(Y,Z)

If X, Y, and Z are independent, or uncorrelated, random

variables, then the covariance terms are zero and:

var(aX bY cZ) = a

2

var(X)

b

2

var(Y) c

2

var(Z)

Normal Probabilities

If X - N(m, s

2

), then Z =

X m

s

-N(0; 1)

If X - N(m, s

2

) and a is a constant, then

P(X _ a) = P Z _

a m

s

_ _

If X -N(m; s

2

) and a and b are constants; then

P(a _ X _ b) = P

am

s

_ Z _

b m

s

_ _

Assumptions of the Simple Linear Regression

Model

SR1 The value of y, for each value of x, is y = b

1

b

2

x e

SR2 The average value of the random error e is

E(e) = 0 since we assume that E(y) = b

1

b

2

x

SR3 The variance of the random error e is var(e) =

s

2

= var(y)

SR4 The covariance between any pair of random

errors, e

i

and e

j

is cov(e

i

, e

j

) = cov(y

i

, y

j

) = 0

SR5 The variable x is not random and must take at

least two different values.

SR6 (optional ) The values of e are normally dis-

tributed about their mean e - N(0, s

2

)

Least Squares Estimation

If b

1

and b

2

are the least squares estimates, then

^y

i

= b

1

b

2

x

i

^e

i

= y

i

^y

i

= y

i

b

1

b

2

x

i

The Normal Equations

Nb

1

Sx

i

b

2

=Sy

i

Sx

i

b

1

Sx

2

i

b

2

= Sx

i

y

i

Least Squares Estimators

b

2

=

S(x

i

x)(y

i

y)

S(x

i

x)

2

b

1

= y b

2

x

Elasticity

h =

percentage change in y

percentage change in x

=

Dy=y

Dx=x

=

Dy

Dx

x

y

h =

DE(y)=E(y)

Dx=x

=

DE(y)

Dx

x

E(y)

= b

2

x

E(y)

Least Squares Expressions Useful for Theory

b

2

= b

2

Sw

i

e

i

w

i

=

x

i

x

S(x

i

x)

2

Sw

i

= 0; Sw

i

x

i

= 1; Sw

2

i

= 1=S(x

i

x)

2

Properties of the Least Squares Estimators

var(b

1

) = s

2

Sx

2

i

NS(x

i

x)

2

_ _

var(b

2

) =

s

2

S(x

i

x)

2

cov(b

1

; b

2

) = s

2

x

S(x

i

x)

2

_ _

Gauss-Markov Theorem: Under the assumptions

SR1SR5 of the linear regression model the estimators

b

1

and b

2

have the smallest variance of all linear and

unbiased estimators of b

1

and b

2

. They are the Best

Linear Unbiased Estimators (BLUE) of b

1

and b

2

.

If we make the normality assumption, assumption

SR6, about the error term, then the least squares esti-

mators are normally distributed.

b

1

- N b

1

;

s

2

x

2

i

NS(x

i

x)

2

_ _

; b

2

- N b

2

;

s

2

S(x

i

x)

2

_ _

Estimated Error Variance

^ s

2

=

S^e

2

i

N 2

Estimator Standard Errors

se(b

1

) =

var(b

1

)

_

; se(b

2

) =

var(b

2

)

_

t-distribution

If assumptions SR1SR6 of the simple linear regression

model hold, then

t =

b

k

b

k

se(b

k

)

- t

(N2)

; k = 1; 2

Interval Estimates

P[b

2

t

c

se(b

2

) _ b

2

_ b

2

t

c

se(b

2

)] = 1 a

Hypothesis Testing

Components of Hypothesis Tests

1. A null hypothesis, H

0

2. An alternative hypothesis, H

1

3. A test statistic

4. A rejection region

5. A conclusion

If the null hypothesis H

0

: b

2

= c is true, then

t =

b

2

c

se(b

2

)

- t

(N2)

Rejection rule for a two-tail test: If the value of the

test statistic falls in the rejection region, either tail of

the t-distribution, then we reject the null hypothesis

and accept the alternative.

Type I error: The null hypothesis is true and we decide

to reject it.

Type II error: The null hypothesis is false and we decide

not to reject it.

p-value rejection rule: When the p-value of a hypoth-

esis test is smaller than the chosen value of a, then the

test procedure leads to rejection of the null hypothesis.

Prediction

y

0

= b

1

b

2

x

0

e

0

; ^y

0

= b

1

b

2

x

0

; f = ^y

0

y

0

var( f ) = ^ s

2

1

1

N

(x

0

x)

2

S(x

i

x)

2

_ _

; se( f ) =

var( f )

_

A (1 a) 100% condence interval, or prediction

interval, for y

0

^y

0

t

c

se( f )

Goodness of Fit

S(y

i

y)

2

= S(^y

i

y)

2

S^e

2

i

SST = SSR SSE

R

2

=

SSR

SST

= 1

SSE

SST

= (corr(y; ^y))

2

Log-Linear Model

ln(y) = b

1

b

2

x e;

ln( y) = b

1

b

2

x

100 b

2

~ % change in y given a one-unit change in x:

^y

n

= exp(b

1

b

2

x)

^y

c

= exp(b

1

b

2

x)exp(^ s

2

=2)

Prediction interval:

exp

ln(y) t

c

se( f )

_ _

; exp

ln( y) t

c

se( f )

_ _

Generalized goodness-of-t measure R

2

g

=(corr(y;^y

n

))

2

Assumptions of the Multiple Regression Model

MR1 y

i

= b

1

b

2

x

i2

b

K

x

iK

e

i

MR2 E(y

i

) =b

1

b

2

x

i2

b

K

x

iK

=E(e

i

) = 0.

MR3 var(y

i

) = var(e

i

) = s

2

MR4 cov(y

i

, y

j

) = cov(e

i

, e

j

) = 0

MR5 The values of x

ik

are not random and are not

exact linear functions of the other explanatory

variables.

MR6 y

i

- N[(b

1

b

2

x

i2

b

K

x

iK

); s

2

[

=e

i

- N(0; s

2

)

Least Squares Estimates in MR Model

Least squares estimates b

1

, b

2

, . . . , b

K

minimize

S(b

1

, b

2

, . . . , b

K

) = (y

i

b

1

b

2

x

i2

b

K

x

iK

)

2

Estimated Error Variance and Estimator

Standard Errors

^ s

2

=

^e

2

i

N K

se(b

k

) =

var(b

k

)

_

Hypothesis Tests and Interval Estimates for Single Parameters

Use t-distribution t =

b

k

b

k

se(b

k

)

- t

(NK)

t-test for More than One Parameter

H

0

: b

2

cb

3

= a

When H

0

is true t =

b

2

cb

3

a

se(b

2

cb

3

)

- t

(NK)

se(b

2

cb

3

) =

var(b

2

) c

2

var(b

3

) 2c

cov(b

2

; b

3

)

_

Joint F-tests

To test J joint hypotheses,

F =

(SSE

R

SSE

U

)=J

SSE

U

=(N K)

To test the overall signicance of the model the null and alternative

hypotheses and F statistic are

H

0

: b

2

= 0; b

3

= 0; : : : ; b

K

= 0

H

1

: at least one of the b

k

is nonzero

F =

(SST SSE)=(K 1)

SSE=(N K)

RESET: A Specication Test

y

i

=b

1

b

2

x

i2

b

3

x

i3

e

i

^y

i

=b

1

b

2

x

i2

b

3

x

i3

y

i

=b

1

b

2

x

i2

b

3

x

i3

g

1

^y

2

i

e

i

; H

0

: g

1

=0

y

i

=b

1

b

2

x

i2

b

3

x

i3

g

1

^y

2

i

g

2

^y

3

i

e

i

; H

0

: g

1

=g

2

=0

Model Selection

AIC = ln(SSE=N) 2K=N

SC = ln(SSE=N) Kln(N)=N

Collinearity and Omitted Variables

y

i

= b

1

b

2

x

i2

b

3

x

i3

e

i

var(b

2

) =

s

2

(1 r

2

23

) (x

i2

x

2

)

2

When x

3

is omitted; bias(b

+

2

) = E(b

+

2

) b

2

= b

3

cov(x

2

; x

3

)

var(x

2

)

Heteroskedasticity

var(y

i

) = var(e

i

) = s

i

2

General variance function

s

2

i

= exp(a

1

a

2

z

i2

a

S

z

iS

)

Breusch-Pagan and White Tests for H

0

: a

2

= a

3

= = a

S

= 0

When H

0

is true x

2

= N R

2

- x

2

(S1)

Goldfeld-Quandt test for H

0

: s

2

M

= s

2

R

versus H

1

: s

2

M

,= s

2

R

When H

0

is true F = ^ s

2

M

=^ s

2

R

- F

(NMKM;NRKR)

Transformed model for var(e

i

) = s

2

i

= s

2

x

i

y

i

=

x

i

_

= b

1

1=

x

i

_

( ) b

2

x

i

=

x

i

_

( ) e

i

=

x

i

_

Estimating the variance function

ln(^e

2

i

) = ln(s

2

i

) v

i

= a

1

a

2

z

i2

a

S

z

iS

v

i

Grouped data

var(e

i

) = s

2

i

=

s

2

M

i = 1; 2; . . . ; N

M

s

2

R

i = 1; 2; . . . ; N

R

_

Transformed model for feasible generalized least squares

y

i

_

^ s

i

_

= b

1

1

_

^ s

i

_

_ _

b

2

x

i

_

^ s

i

_

_ _

e

i

_

^ s

i

_

Regression with Stationary Time Series Variables

Finite distributed lag model

y

t

=a b

0

x

t

b

1

x

t1

b

2

x

t2

b

q

x

tq

v

t

Correlogram

r

k

= (y

t

y)(y

tk

y)= (y

t

y)

2

For H

0

: r

k

= 0; z =

T

_

r

k

- N(0; 1)

LM test

y

t

=b

1

b

2

x

t

r^e

t1

^v

t

Test H

0

: r =0 with t-test

^e

t

=g

1

g

2

x

t

r^e

t1

^v

t

Test using LM=T R

2

AR(1) error y

t

=b

1

b

2

x

t

e

t

e

t

= re

t1

v

t

Nonlinear least squares estimation

y

t

= b

1

(1 r) b

2

x

t

ry

t1

b

2

rx

t1

v

t

ARDL(p, q) model

y

t

= d d

0

x

t

d

l

x

t1

d

q

x

tq

u

l

y

t1

u

p

y

tp

v

t

AR(p) forecasting model

y

t

= d u

l

y

t1

u

2

y

t2

u

p

y

tp

v

t

Exponential smoothing ^y

t

= ay

t1

(1 a)^y

t1

Multiplier analysis

d

0

d

1

L d

2

L

2

d

q

L

q

= (1 u

1

L u

2

L

2

u

p

L

p

)

(b

0

b

1

L b

2

L

2

)

Unit Roots and Cointegration

Unit Root Test for Stationarity: Null hypothesis:

H

0

: g = 0

Dickey-Fuller Test 1 (no constant and no trend):

Dy

t

= gy

t1

v

t

Dickey-Fuller Test 2 (with constant but no trend):

Dy

t

= a gy

t1

v

t

Dickey-Fuller Test 3 (with constant and with trend):

Dy

t

= a gy

t1

lt v

t

Augmented Dickey-Fuller Tests:

Dy

t

= a gy

t1

m

s=1

a

s

Dy

ts

v

t

Test for cointegration

D^e

t

= g^e

t1

v

t

Random walk: y

t

= y

t1

v

t

Random walk with drift: y

t

= a y

t1

v

t

Random walk model with drift and time trend:

y

t

= a dt y

t1

v

t

Panel Data

Pooled least squares regression

y

it

= b

1

b

2

x

2it

b

3

x

3it

e

it

Cluster robust standard errors cov(e

it

, e

is

) = c

ts

Fixed effects model

y

it

= b

1i

b

2

x

2it

b

3

x

3it

e

it

b

1i

not random

y

it

y

i

= b

2

(x

2it

x

2i

) b

3

(x

3it

x

3i

) (e

it

e

i

)

Random effects model

y

it

=b

1i

b

2

x

2it

b

3

x

3it

e

it

b

it

=b

1

u

i

random

y

it

ay

i

=b

1

(1a) b

2

(x

2it

ax

2i

) b

3

(x

3it

ax

3i

) v

+

it

a=1s

e

_

Ts

2

u

s

2

e

_

Hausman test

t = (b

FE;k

b

RE;k

)

_

var(b

FE;k

)

var(b

RE;k

)

_ _

1=2

- Application-of-Artificial-Intelligence-Approach-in-Modeling-_2016_Procedia-T.pdfUploaded byTanNguyễn
- Solutions Manual Ch13 - 2012Uploaded bythegreatll
- DeVry MATH 533 Final Exam 100% Correct AnswerUploaded byDeVryHelp
- Letter to ShareholdersUploaded byKoral Li
- Week 2 Individual y GrupalUploaded byMaria Esther Feliciano Delgado
- Standard CostingUploaded byrrkabra
- 143143458 Brand LoyaltyUploaded byJam Imran Madni
- Chapter 16Uploaded byNdomadu
- STATISTICS IN EDUCATION.pdfUploaded byZoe Teng
- Literacy Practices and SchoolingUploaded byCarlaRein
- 2 - Bowman Helfat - Does Corporate Strategy Matter (1)Uploaded byShirinda Pradeepti
- Manova x2 x19 x20Uploaded byby21
- Stats Ch8 FormulasUploaded byPoonam Naidu
- jurnal honey comb.pdfUploaded byDharma Pemuraz
- qtec-35-4-335Uploaded byPhdKaizer
- The Evaluation Relationship Between the Factors Affecting the Security of Information Systems and the Financial ReportingUploaded byMAGNT Research Report
- initial reportUploaded byapi-314733073
- quality manajemenUploaded byAdi Irawan
- research paper samstipicUploaded byapi-273337572
- 180Uploaded bygotomaster
- kuesioner validasi 0.5.docUploaded bypopokyu
- CMAES 2Uploaded byMiguel Lopez
- Queuing Analytic Theory and Discrete Events Simulation.pdfUploaded byAbiyyah Wiriana
- UTILITY OF “A SYSTEMATIC ALPHABETICAL REPERTORY OF HOMOEOPATHIC REMEDIES BY DR. C. VON BOENNINGHAUSEN IN THE CASES OF ACNE VULGARIS” - A RANDOMIZED SINGLE BLIND CONTROL TRIALUploaded byAnonymous CwJeBCAXp
- Hyper Geometric DistributionUploaded bynovati8x
- C_7_03Uploaded bysatchvn
- MB0034Uploaded byVenu Gopal
- 20140303_Test1.pdfUploaded by2009028818
- Newsworthy SurveyUploaded byjuliecmckinney
- Notes on Distance Dependence of Pair Correlation StatisticsUploaded byHtemae

- MulticollinearityUploaded bymissinu
- Stata RUploaded bymissinu
- Exam ECON301B 2002 CommentedUploaded bymissinu
- Exam ECON301B 2002 CommentedUploaded bymissinu
- AnkiUploaded bymissinu
- Lecture 1 - Overview of Supervised LearningUploaded bymissinu
- Exit Questionnaire 2005 FinalUploaded bymissinu
- Economic Accounts 2005 FinalUploaded bymissinu
- Economic Accounts 2005 FinalUploaded bymissinu
- Exam4135 2004 SolutionsUploaded bymissinu
- Lecture 4 - Basis Expansion and RegularizationUploaded bymissinu
- Balabolka SampleUploaded bymissinu
- Bai Giang Toan C2 (2009)Uploaded bymissinu
- Lecture 2 - Some Course AdminUploaded bymissinu
- Lecture 3 - Linear Methods for ClassificationUploaded bymissinu
- Cuc BVTVUploaded bymissinu
- Midterm Microeconomics 1 2012-13Uploaded bymissinu
- Thuchanh CH 141030Uploaded bymissinu
- Bai Giang Toan Kinh Te Quang 2012 1171Uploaded bynicksforums
- Giao Trinh VBA_GXDUploaded byYumi Ling
- Bai Tap Giai Tich 2 Chuong 2Uploaded bymissinu
- Tom Tat Cong Thuc XSTKUploaded byTuấn Lê
- Regulations Livestock in VN SummaryUploaded bymissinu
- Manure Estimates.pdfUploaded bymissinu
- Introduction to Microeconomic Theory and GE Theory (2015)Uploaded bymissinu
- 3-Vu Trong Khai - Tich Tu Ruong DatUploaded byematn
- GRE VocabularyUploaded byKoksiong Poon
- Visual BasicUploaded byxuananh
- Giai Tich 2 2014 Chuong 5Uploaded bymissinu
- Vocabulary IELTS Speaking Theo TopicUploaded byBBBBBBB

- Undergraduate EconometricUploaded byAcho Jie
- dynareUploaded byJoab Dan Valdivia Coria
- Guide to Modern EconometricsUploaded byPuspichanPalazzo
- Manuale SuperAmpelo - En DraftUploaded byJelena Koković
- Indian Stock Market and Great RecessionUploaded bysiddula
- Evaluation of Venture Capital Based on Evaluation ModelUploaded bykamer4u
- An Empirical Study of the Relationships between CO2 Emissions, Economic Growth and Openness.pdfUploaded bySiddharth Venkataraman
- Bacteria Charles RiverUploaded byYaasiin Oozeer
- Adkins (2011). Using gretl for Principles of Econometrics, 4th Edition.pdfUploaded byWilliamtom
- Causality in Economics and EconometricsUploaded byAbigail P. Dumalus
- 552_Notes_6aUploaded bycfisicaster
- Effects of Word-Of-Mouth Versus Traditional Marketing Findings From an Internet SocialUploaded byJavier Gonzalez
- Univariate Time Series Part IUploaded byFeña Pino Magna
- Tutorial on VARUploaded bysom
- Ardl ModelUploaded bysakiaslam
- Logistic Regression sample size calculationUploaded byAmado Saavedra
- An Examination of Blujme and Vasicek BetasUploaded byblindslayer
- Solomon L QP - S1 EdexcelUploaded byabhay
- Stage & Crookston (2007)Uploaded byRock e Lio
- Chap 12Uploaded byAmsalu Walelign
- CoInte - Johansen (1992)Uploaded byAlexander Baldeon Arrunategui
- c2Uploaded bySheriad Majie
- Bull Wip EffectUploaded byykbharti101
- BDLagM-BiometricsUploaded byrdpeng
- Andrew Tulloch - ESL-SolutionsUploaded byAnonymous901
- KayaUploaded byMichael Evans
- Why was Japan hit so hard by the Financial Crisis.pdfUploaded byLewis
- 12 3 2010 Mitchel SlidesUploaded byHector Garcia
- Bài Price Puzzle Đầu TiênUploaded bytoilanhan1977
- Poi (2002)Uploaded bynash0612