8 views

Uploaded by teikkheong

INtro to CFA SEM

- 173232298 a Guide to Modern Econometrics by Verbeek 151 160
- Cox Regression Methods
- WP_81_2007_Bovi
- spss 簡單線性回歸分析 spss simple regression.pdf
- Enhancement of Laccase Production from Pleurotus ostreatus PVCRSP-7 by altering the Nutritional Conditions using Response Surface Methodology
- Analysis of Formal Strategic Planning and Firm Performance
- Rosso1995 Effect PH T
- 13. Huawei - Chinese Telecommunications Giant Huawei- Strategies to Success
- Saint Gba334 Module 3 Quiz 2
- Analysis of Residuals — a Useful Tool for Phase Equilibrium Data Analysis
- Eviews_Commandref40
- Bum2413 - Applied Statistics 21112
- Jurnal Data Primer
- Regression Explained in Simple Terms
- Assignment 3
- Groups With Prob 2009-06-25
- Exercises Dobson
- Goldstein, J. P. (1999). Predator-Prey Model Estimates of the Cyclical Profit Squeeze. Metroeconomica 50, 2, pp. 139-173.pdf
- Forecasting.ppt
- International Refereed Journal of Engineering and Science (IRJES)

You are on page 1of 35

BUSI6280

The term structural equation modeling

conveys two key aspects of the procedure:

That the causal processes under study are

represented by a series of structural (i.e.,

regression) equations,

That these structural relations can be

modeled pictorially to enable a clearer

conceptualization of the theory under

study.

Why Use SEM ?

SEM lends itself well to the analysis of data for

inferential purposes.

Whereas, traditional multivariate procedures are

incapable of either assessing or correcting for

measurement error, SEM provides explicit

estimates of these parameters.

SEM procedures can incorporate both unobserved

(i.e. latent) and observed variables.

The factor analytic model (EFA or CFA) focuses solely

on how the observed variables are linked to their

underlying latent factors.

Factor analysis is concerned with the extent to which

the observed variables are generated by the underlying

latent constructs and thus strength of the regression paths

from the factors to the observed variables (the factor

loadings) are of primary interest.

Although inter-factor relations are also of interest, any

regression structure among them is not considered in the

factor analytic model.

Purpose of Factor Analysis

Measurement model: latent variables and their observed

measures (i.e., the CFA model)

Structural model: Model with links among the latent

variables.

Full (Complete) Model: a measurement model and a

structural model

Recursive model: Direction of cause is from one direction

only

Non-recursive model: reciprocal or feedback effects (often

different from one another).

Type of Models

Three different scenarios or models

(Jreskog 1993)

Strictly Confirmatory (SC)

Alternative Models (AM)

Model Generating (MG)

The SC Scenario

The researcher postulates a single model

based on theory, collects the appropriate

data, and then tests the fit of the

hypothesized model to the sample data.

From the results of this test, the researcher

either rejects or fails to reject the model. No

further modifications to the model are made.

The AM Scenario

The researcher proposes several alternative

(i.e., competing) models, all of which are

grounded in theory.

Following analysis of a single set of empirical

data, the researcher selects one model as

most appropriate in representing the sample

data.

The MG Scenario

The researcher, having postulated and rejected a

theoretically derived model on the basis of its poor fit to

the sample data, proceeds in an exploratory (rather than

confirmatory) fashion to modify and re-estimate the

model.

The primary focus here is to locate the source of misfit

in the model. Jreskog noted that, although

respecification may be either theory- or data-driven, the

ultimate objective is to find a model that is both substantively

meaningful and statistically well fitting.

SEM procedures alternative computer programs

AMOS-Arbuckle, 1995

EQS-Bentler, 1995

LISCOMP-Muthn, 1998

CALIS-SAS Institute, 1992

RAMONA-Browne, Mels, & Coward, 1994

SEPATH-Steiger, 1994

LISREL program is the most widely used, 1970s

Exogenous latent variables are synonymous with independent

variables; they cause fluctuations in the values of other

latent variables in the model.

Changes in the values of exogenous variables are not

explained by the model. Rather, they are considered to be

influenced by other factors external to the model.

Endogenous latent variables are synonymous with dependent

variables and, as such, are influenced by the exogenous

variables in the model, either directly, or indirectly.

SEM - Language

By convention, observed measures are represented by Roman

letters and latent constructs by Greek letters:

Those that are exogenous are termed X-variables.

Those that are endogenous are termed Y-variables.

The measurement model may be specified either in terms

of LISREL exogenous notation (i.e., X-variables), or in

terms of its endogenous notation (i.e., Y-variables).

The exogenous latent constructs are termed as (xi).

The endogenous latent constructs are termed as (eta).

SEM - Language

SEM Language

The Measurement Model

x is a q x 1 vector of observed exogenous variables

y is a p x 1 vector of observed endogenous variables.

is an n x 1 vector of latent exogenous variables

is an m x 1 vector of latent endogenous variables.

is a q x n matrix of coefficients (ij) linking x and

.

is a q x 1 vector of random disturbance term

(errors of measurement) associated with x vector.

is a p x 1 vector of random disturbance term

(errors of measurement) associated with y vector.

SEM Language

The Structural Model

(gamma) is an m x n matrix of coefficients (ij) that

relates the n exogenous factors to the m endogenous

factors.

B(beta) is an m x m matrix of coefficients (ij) that relates

the m endogenous factors to one another.

(zeta) is an m x 1 vector of residuals (i) representing

errors in the equation relating and .

(phi) is an n x n matrix of coefficients (ij) that

captures the variance/covariance between s.

(psi) is the m x m matrix of covariance between s.

The Structural Model

Measurement Model for the X-variables (1):

x=

x

+

Measurement Model for the Y-variables (2):

y=

y

+

Structural Equation Model (3):

=B + +

The following minimal assumptions are

presumed to hold for the system of equations

is uncorrelated with (construct)

is uncorrelated with (construct)

is uncorrelated with and (construct)

, , and are mutually uncorrelated.

E() = 0

E() = 0

E() = 0

E() = 0

E() = 0

(I-B) is nonsingular so that (I-B) exists. This makes the

equation 3 to be written in the reduced form.

Symbol

Representation

Unobserved (latent) Factors

Observed Variable

Path coefficient for regression of

observed variable on unobserved

factors

Path coefficient for regression of

one factor on another.

Residual error (disturbance) in

prediction of unobserved factors

Measurement error associated

with observed variable.

Summary of Matrices, Greek Notation, and Programs Codes

Matrix Program Matrix

Greek Letter Matrix Element Code Type

Measurement Model

Lambda-X

x

x

LX

Regression

Lambda-Y

y

y

LY Regression

Theta delta Q

TD Var/cov

Theta epsilon Q

e

e

TE Var/cov

Structural Model

Gamma GA Regression

Beta BE Regression

Phi PH Var/cov

Psi PS Var/cov

Xi (or Ksi) --- --- Vector

Eta --- --- Vector

Zeta --- --- Vector

Var/cov = variance-covariance

The Structural Model -

1

predicted by

1

x11

x21

x31

1

X

1

X

2

X

3

1

Y

1

Y

2

y11

y21

An important corollary of SEM is that the

variances and covariance of dependent (or

endogenous) variables, whether they be

observed or unobserved, are never

parameters of the model; these are

explained by the exogenous variables.

In contrast, the variance and covariance of

independent variable are important

parameters that need to be estimated.

MEASUREMENT (CFA) MODELS

1

X

1

11

2

X

2

21

1

3

X

3

31

CFA Part

1

1

11

Y

1

1

21

Y

2

2

CFA Part

CFA Model

error ReadSC

ASC

error WriteSC

error TalkSC

SSC

error InteractSC

CFA with Greek Notation

1

x

1

11

1

2

x

2

11

21

3

x

3

32

2

4

x

4

42

Regression Equations (Xs)

1. x

1

=

11

1

+

1

2. x

2

=

21

1

+

2

3. x

3

=

32

2

+

3

4. x

4

=

42

2

+

4

Or in matrix form

X =

x

+

The parameters of this model are

x

,

,and

Where:

x

represents the matrix of regression

coefficients related to the s (described earlier).

(phi) is an x symmetrical variance-

covariance matrix among the exogenous

factors.

(theta-delta) is a symmetrical q x q variance-

covariance matrix among the error of

measurement for the q exogenous observed

variables

The general factor analytic model

can be expanded as:

X =

x

+

x

1

11

0

1

x

2

21

0

1

2

= +

x

3

0

32

2

3

x

4

0

42

4

is the Loadings Matrix

The matrix is often termed the factor-

loading matrix because it portrays the

pattern by which each observed variable is

linked to its respective factor.

The Ys

1. y

1

=

11

1

+

1

2. y

2

=

21

1

+

2

3. y

3

=

32

2

+

3

4. y

4

=

42

2

+

4

Or in matrix form

Y =

y

+

Matrix Notation for Loadings

with Regression Model

Y =

y

+

y

1

11

0

1

y

2 =

21

0

1

+

2

y

3

0

32

2

3

y

4

0

32

4

A just-identified model is one in which

there is a one-to-one correspondence between the

data and the structural parameters.

Number of data variances and covariances equal

number of parameters to be estimated.

However, despite the capability of the model to

yield a unique solution for all parameters, the just-

identified model is not scientifically interesting

because it has no degrees of freedom and

therefore can never be rejected.

Overidentified Model

An overidentified model is one in which the

number of estimable parameters is less than the

number of data points (i.e., variance, covariance of

the observed variable).

This situation results in positive degrees of

freedom that allows for rejection of the model,

thereby rendering it scientific use. The aim in

SEM, then, is to specify a model such that it

meets the criterion of overidentification.

Underidentified Model

An underidentified model is one in which the

number of parameters to be estimated exceeds the

number of variances and covariances.

As such, the model contains insufficient

information (from the input data) for the

purpose of attaining a determinate solution of

parameter estimation; that is, an infinite number

of solutions are possible for an underidentified

model.

Suppose there are 12 observed variable, this means that

we have 12(12+ 1)/2=78 data points.

Suppose that there are 30 unknown parameters.

Thus, with 78 data points and 30 parameters to be

estimated, we have an overidentified model with 48

degrees of freedom.

It is important to point out, however, that the

specification of an overidentified model is a

necessary, but not sufficient condition to resolve the

identification problem. Indeed, the imposition of

constraints on particular parameters can sometime be

beneficial in helping the researcher to attain an

overidentified model

No Scale Set for Constructs

Linked to the issue of identification is the

requirement that every latent variable have

its scale determined. This requirement

arises because these variable are unobserved

and therefore have no definite metric scale;

Assume CFA Model with 12 variables (items) and

4 factors (3 items per factor).

We can assume that there are 12 regression coefficient (s)

There are 12 error variance (s).

There are 4 factors variances (which may be standardized

and therefore set to 1).

There are 6 covariances between factors.

If the factor variances are not set to 1 then then one of the

parameters for each factor can be fixed to a value of

1.00 (they are therefore not to be estimated). The rationale

underlying this constraint is tied to the issue of statistical

identification. In total, then, there are 30 parameters to be

estimated for this CFA model.

- 173232298 a Guide to Modern Econometrics by Verbeek 151 160Uploaded byAnonymous T2LhplU
- Cox Regression MethodsUploaded bySzu-Yu Kao
- WP_81_2007_BoviUploaded byhalhoshan
- spss 簡單線性回歸分析 spss simple regression.pdfUploaded byvictor_chung_23
- Enhancement of Laccase Production from Pleurotus ostreatus PVCRSP-7 by altering the Nutritional Conditions using Response Surface MethodologyUploaded byDr. Thadikamala Sathish
- Analysis of Formal Strategic Planning and Firm PerformanceUploaded byRohit Sinha
- Rosso1995 Effect PH TUploaded bymisael
- 13. Huawei - Chinese Telecommunications Giant Huawei- Strategies to SuccessUploaded byRamya Gunukula
- Saint Gba334 Module 3 Quiz 2Uploaded byteacher.theacestud
- Analysis of Residuals — a Useful Tool for Phase Equilibrium Data AnalysisUploaded byMario Ricardo Urdaneta Parra
- Eviews_Commandref40Uploaded byCatalin Lefter
- Bum2413 - Applied Statistics 21112Uploaded bySyada Hageda
- Jurnal Data PrimerUploaded byanto
- Regression Explained in Simple TermsUploaded byryankosk
- Assignment 3Uploaded byshagunparmar
- Groups With Prob 2009-06-25Uploaded byAnabelle Belizario
- Exercises DobsonUploaded byvsuarezf2732
- Goldstein, J. P. (1999). Predator-Prey Model Estimates of the Cyclical Profit Squeeze. Metroeconomica 50, 2, pp. 139-173.pdfUploaded bylcr89
- Forecasting.pptUploaded bypkkavithaa
- International Refereed Journal of Engineering and Science (IRJES)Uploaded bywww.irjes.com
- The People's Republic of China as an Engine of Growth for Developing Asia?: Evidence from Vector Autoregression ModelsUploaded byAsian Development Bank
- Book1Uploaded byJojo
- marketing research ppt 1 1Uploaded byapi-292900866
- NumXL FunctionsUploaded bySpider Financial
- Data Mining in Excel Using Xl MinerUploaded bychintan18
- SA3 Assignment 1Uploaded byPrasanna Kumar Mamidala
- Phenological Models of Flower Bud Stages and Frui Growth of Montmorency Sour CherryUploaded byAlvaro Andres
- Multivar 2 - Simple and Multiple Regression.pdfUploaded byokisaputra198909
- Data.xlsxUploaded byαγαπημένη του Χριστού
- SPSS- how toUploaded byNikhil Sharma

- Bayesian Methods in FinanceUploaded byteikkheong
- Volatility ForecastingUploaded byDino Mandic
- Bayesian Portfolio AnalysisUploaded byteikkheong
- Topic 5 Research DesignUploaded byteikkheong
- HarvardUploaded byteikkheong
- 160515-A Bayesian Approach to Implied Volatility Analysis of s&p 500Uploaded byteikkheong
- Intro to ResearchUploaded byteikkheong
- Evolution of Market Uncertainty Around EarningsUploaded byteikkheong
- Heterogeneous Beliefs and the Option-implied Volatility SmileUploaded byteikkheong
- Options StrategiesUploaded byteikkheong
- Contributions to Bayesian Structural EquationUploaded byteikkheong
- BayesianUploaded byteikkheong
- AMOS TutorialUploaded byPolksjh Sdjqlw
- Dubinsky Earnings Annuncement and Equity Options - Working Paper 2005Uploaded byteikkheong

- 04 Percntiles Deciles and QuartilesUploaded byKathleen
- Hypothesis TestingUploaded byshomitamaraewu
- STHELPUploaded byNeha_ Sharma08
- asg1solnUploaded bylillyren2808
- Unit 3Uploaded bylghmshari
- Financial Time Series AnalysisUploaded byEmmanuel K'pkor
- Generalized Linear ModelsUploaded byJosé Manuel Carranza
- Martin Hall Descriptive and Inferential Statistics Write UpUploaded bymartin_tim_hall7434
- 255185Uploaded byangeljosechuquiure
- Stat130 assignment 4 on Unit_5.docxUploaded byUsman
- I PTUploaded byAnupam Saxena
- Mle Mini Lecture NotesUploaded by2456903
- Quantitative MethodsUploaded byJoseph Anbarasu
- CCP303Uploaded byapi-3849444
- kedd.pdfUploaded byMei Fadillah
- Error analysis in diameter estimationUploaded byLaura Ileana Stoilescu
- Control ChartsUploaded byAkash Kazama
- MH3511midterm2017Q (1)Uploaded byFrancis Tan
- Basic Probability and Statistics Review Six Sigma Black Belt PrimerUploaded byJosé Esqueda Leyva
- Mh 3511 Midterm 2018 So LnUploaded byFrancis Tan
- AP Statistics 1st Semester Study GuideUploaded bySusan Huynh
- Lesson 05 Forecasting & Smoothing MethodsUploaded byCarl Bennett
- Hayashi EconometricsUploaded byateranimus
- Probability ExercisesUploaded byAnu Uy
- Basic Statistics for Health SciencesUploaded bygeofflilley
- HW8Uploaded byPallavi Raiturkar
- Assignment 1Uploaded bysnazrul
- MidSem-2013S2Uploaded byHsingliang Tay
- process based quality management system.docxUploaded byselinasimpson321
- Cinical Global Impression of Severity for Schizoaffective Disorder Scale and Established Mood Scale for ManiaUploaded byDivaviya