Title: Factor Analysis and Inference for Structured Covariance Matrices
1Factor Analysis and Inference for Structured
Covariance Matrices
- Shyh-Kang Jeng
- Department of Electrical Engineering/
- Graduate Institute of Communication/
- Graduate Institute of Networking and Multimedia
2History
- Early 20th-century attempt to define and measure
intelligence - Developed primarily by scientists interested in
psychometrics - Advent of computers generated a renewed interest
- Each application must be examined on its own
merits
3Essence of Factor Analysis
- Describe the covariance among many variables in
terms of a few underlying, but unobservable,
random factors. - A group of variables highly correlated among
themselves, but having relatively small
correlations with variables in different groups
represent a single underlying factor
4Example 9.8Examination Scores
5Orthogonal Factor Model
6Orthogonal Factor Model
7Orthogonal Factor Model
8Orthogonal Factor Model
9Orthogonal Factor Model
10Example 9.1 Verification
11Example 9.2 No Solution
12Ambiguities of L When mgt1
13Principal Component Solution
14Principal Component Solution
15Residual Matrix
16Determination of Number of Common Factors
17Example 9.3Consumer Preference Data
18Example 9.3Determination of m
19Example 9.3Principal Component Solution
20Example 9.3Factorization
21Example 9.4Stock Price Data
- Weekly rates of return for five stocks
- X1 Allied Chemical
- X2 du Pont
- X3 Union Carbide
- X4 Exxon
- X5 Texaco
22Example 9.4Stock Price Data
23Example 9.4Principal Component Solution
24Example 9.4Residual Matrix for m2
25Maximum Likelihood Method
26Result 9.1
27Factorization of R
28Example 9.5 Factorization ofStock Price Data
29Example 9.5ML Residual Matrix
30Example 9.6Olympic Decathlon Data
31Example 9.6Factorization
32Example 9.6PC Residual Matrix
33Example 9.6ML Residual Matrix
34A Large Sample Test for Number of Common Factors
35A Large Sample Test for Number of Common Factors
36Example 9.7Stock Price Model Testing
37Example 9.8Examination Scores
38Example 9.8Maximum Likelihood Solution
39Example 9.8Factor Rotation
40Example 9.8Rotated Factor Loading
41Varimax Criterion
42Example 9.9 Consumer-Preference Factor Analysis
43Example 9.9Factor Rotation
44Example 9.10 Stock Price Factor Analysis
45Example 9.11Olympic Decathlon Factor Analysis
46Example 9.11Rotated ML Loadings
47Factor Scores
48Weighted Least Squares Method
49Factor Scores of Principal Component Method
50Orthogonal Factor Model
51Regression Model
52Factor Scores by Regression
53Example 9.12Stock Price Data
54Example 9.12Factor Scores by Regression
55Example 9.13 Simple Summary Scores for Stock
Price Data
56A Strategy for Factor Analysis
- 1. Perform a principal component factor analysis
- Look for suspicious observations by plotting the
factor scores - Try a varimax rotation
- 2. Perform a maximum likelihood factor analysis,
including a varimax rotation
57A Strategy for Factor Analysis
- 3. Compare the solutions obtained from the two
factor analyses - Do the loadings group in the same manner?
- Plot factor scores obtained for PC against scores
from ML analysis - 4. Repeat the first 3 steps for other numbers of
common factors - 5. For large data sets, split them in half and
perform factor analysis on each part. Compare the
two results with each other and with that from
the complete data set
58Example 9.14Chicken-Bone Data
59Example 9.14Principal Component Factor Analysis
Results
60Example 9.14 Maximum Likelihood Factor Analysis
Results
61Example 9.14Residual Matrix for ML Estimates
62Example 9.14Factor Scores for Factors 1 2
63Example 9.14Pairs of Factor Scores Factor 1
64Example 9.14Pairs of Factor Scores Factor 2
65Example 9.14Pairs of Factor Scores Factor 3
66Example 9.14Divided Data Set
67Example 9.14 PC Factor Analysis for Divided Data
Set
68WOW Criterion
- In practice the vast majority of attempted factor
analyses do not yield clear-cut results - If, while scrutinizing the factor analysis, the
investigator can shout Wow, I understand these
factors, the application is deemed successful
69Structural Equation Models
- Sets of linear equations to specify phenomena in
terms of their presumed cause-and-effect
variables - In its most general form, the models allow for
variables that can not be measured directly - Particularly helpful in the social and behavioral
science
70LISREL (Linear Structural Relationships) Model
71Example
- h performance of the firm
- x managerial talent
- Y1 profit
- Y2 common stock price
- X1 years of chief executive experience
- X2 memberships on board of directors
72Linear System in Control Theory
73Kinds of Variables
- Exogenous variables not influenced by other
variables in the system - Endogenous variables affected by other variables
- Residual associated with each of the dependent
variables
74Construction of a Path Diagram
- Straight arrow
- to each dependent (endogenous) variable from each
of its source - Straight arrow
- also to each dependent variables from its
residual - Curved, double-headed arrow
- between each pair of independent (exogenous)
variables thought to have nonzero correlation
75Example 9.15Path Diagram
76Example 9.15Structural Equation
77Covariance Structure
78Estimation
79Example 9.16Artificial Data
80Example 9.16Artificial Data
81Example 9.16Artificial Data
82Assessing the Fit of the Model
- The number of observations p for Y and q for X
must be larger than the total number of unknown
parameters t lt (pq)(pq1)/2 - Parameter estimates should have appropriate signs
and magnitudes - Entries in the residual matrix S S should be
uniformly small
83Model-Fitting Strategy
- Generate parameter estimates using several
criteria and compare the estimates - Are signs and magnitudes consistent?
- Are all variance estimates positive?
- Are the residual matrices similar?
- Do the analysis with both S and R
- Split large data sets in half and perform the
analysis on each half