Soc 63993, Advanced Social Statistics II
Homework No. 9
Logistic Regression
I.
As we saw in the class handout on the PSI teaching example, 8 of the 14 students who
were in PSI got As compared to only 3 of the 18 students who were in a conventional classro
Interaction effects between continuous variables
This is a very brief overview of this somewhat difficult topic. The course readings provide much
more detail, and you should go over these carefully if you feel these kinds of interaction terms
will be usef
Discussion questions for group comparisons and interaction effects
If time permits we will go over these in class or lab; otherwise just think about them on your own.
1.
The notes present several models for interaction effects.
a.
Suppose your dependent v
Interpreting Interaction Effects; Interaction Effects and Centering
Models with interaction effects can be a little confusing to understand. The handout provides
further discussion of how interaction terms should be interpreted and how centering continuou
Models for Group Comparisons Summary
Since we are estimating and comparing several models, it will be helpful to list several of them
all in one place. This handout summarizes how to do group comparisons both by running
separate models for each group and
Interaction effects and group comparisons
Alternative strategy for testing whether parameters differ across groups: Dummy
variables and interaction terms. We have previously shown how to do a global test of whether
any coefficients differ across groups. T
Group Comparisons:
Using What If Scenarios to Decompose Differences Across Groups
We saw that the effects of education and job experience were smaller for blacks than for whites.
However, we also saw that blacks had lower levels of education and less job
Group Comparisons:
Differences in Composition Versus Differences in Models and Effects
Overview. This is the first of a series of handouts that will deal with techniques for comparing
groups. This initial handout notes that, when comparing groups, it is i
Specification Error: Omitted and Extraneous Variables
Omitted variable bias. Suppose that the correct model is
y 1 X1 2 X 2
If we estimate
y a b1 X1 b2 X 2 e
we know that E(b1) = 1 and E(b2) = 2 i.e. the regression coefficients are unbiased estimators of
Serial Correlation (Very Brief Overview)
[NOTE: These notes draw very, very heavily from Pindyck and Rubinfeld.]
Introduction. Ive never had much reason to worry about serial correlation in my work. But, if
you are working with data that are collected rep
Heteroscedasticity
[NOTE: These notes draw heavily from Berry and Feldman, and, to a lesser extent, Allison, and
Pindyck and Rubinfeld.]
What heteroscedasticity is. Recall that OLS makes the assumption that
V ( j ) 2 for all j. That is, the variance of th
Outliers
[NOTE: These notes draw heavily from several sources, including Foxs Regression Diagnostics;
Pindyck and Rubinfeld; Statistics for Social Data Analysis, by George Bohrnstedt and David
Knoke, 1982; Norusiss SPSS 11 chapter 22 on Analyzing residual
Intro to path analysis
Sources. This discussion draws heavily from Otis Dudley Duncans Introduction to Structural
Equation Models.
Overview. Our theories often lead us to be interested in how a series of variables are interrelated.
It is therefore often d
Intro to Path Analysis - Highlights
Sample Model
u
X2
X1
X4
w
X3
v
Structural Equations for the Above Model
X 2 21 X 1 u
X 3 31 X 1 32 X 2 v
X 4 41 X 1 42 X 2 43 X 3 w
Correlations implied by the model (if all variables are in standardized form)
21 21
31
Structural Coefficients in Recursive Models/ Evils of Standardization
RECURSIVE DEFINED. A model is said to be recursive if all the causal linkages run one way,
that is, no two variables are reciprocally related in such a way that each affects and depends
Course Syllabus for Sociology 63993
Graduate Statistics II
Spring 2011
Instructor
Richard Williams
741 Flanner (Office: 574-631-6668, Cell: 574-360-1017)
Office Hours: MW 10:45-11:30 and by appointment
Immediately before & after class is also good.
If you
Using Stata 9 & Higher for OLS Regression
Introduction. Stata is a popular alternative to SPSS, especially for more advanced statistical
techniques. This handout shows you how Stata can be used for OLS regression. It assumes
knowledge of the statistical c
Using Stata 9 & 10 for Logistic Regression
NOTE: The routines spost9, lrdrop1, and extremes are used in this handout. Use the
findit command to locate and install them. See related handouts for the statistical theory
underlying logistic regression and for
Brief Overview of LISREL & Related Programs & Techniques
STRUCTURAL AND MEASUREMENT MODELS: LISREL. We have focused on structural models.
Such models assume that all variables are measured without error. Of course, this assumption is
often not reasonable.
Brief Overview of Manova
In this and other handouts, well briefly go over some advanced techniques that can be useful
when estimating complicated models. We wont discuss these in detail, but at least youll know
what to look up should you encounter such pr
Nonrecursive models
[NOTE: This lecture borrows heavily from Duncans Introduction to Structural Equation Models
and from William D. Berrys Nonrecursive Causal Models.
Advantages and Disadvantages of Recursive Models. We have previously considered
recursiv
Multinomial Logit Models - Overview
This is adapted heavily from Menards Applied Logistic Regression analysis; also, Borooahs
Logit and Probit: Ordered and Multinomial Models; Also, Hamiltons Statistics with Stata,
Updated for Version 7.
When categories a
Ordered Logit Models - Overview
This is adapted heavily from Menards Applied Logistic Regression analysis; also, Borooahs
Logit and Probit: Ordered and Multinomial Models; Also, Hamiltons Statistics with Stata,
Updated for Version 7.
We have talked about
Logistic Regression, Part III:
Hypothesis Testing, Comparisons to OLS
[This handout steals heavily from the SPSS Advanced Statistics User Guide. Also, Linear
probability, logit, and probit models, by John Aldrich and Forrest Nelson, paper # 45 in the Sage
Logistic Regression, Part II:
The Logistic Regression Model (LRM) Interpreting Parameters
[This handout steals heavily from Linear probability, logit, and probit models, by John Aldrich
and Forrest Nelson, paper # 45 in the Sage series on Quantitative App
Logistic Regression, Part I:
Problems with the Linear Probability Model (LPM)
[This handout steals heavily from Linear probability, logit, and probit models, by John Aldrich
and Forrest Nelson, paper # 45 in the Sage series on Quantitative Applications in
Computing R2/Evils of R2
COMPUTING R2. Here are some of the many formulas for R2: Our knowledge of path analysis now
makes it possible to prove many of these formulas. See the optional appendix if you are
interested. NOTE: I use the notation bk for the st
Scale Construction (Very Brief Overview)
This handout draws heavily from Marija Norusiss SPSS 14.0 Statistical Procedures Companion. See her
chapters 18 (Reliability Analysis) and 17 (Factor Analysis), as well as Hamiltons (2004) ch. 12
(Principal Compone
Measurement Error Example (Supplemental)
Here we give a hypothetical example that illustrates the properties shown in the measurement
handout. We create a data set where the true measures (Yt and Xt) have a correlation of .7 with
each other but the observ