POLI 8501
Binary Logit & Probit, I
The LPM, Logit, and Probit. Consider the case of a binary response (dependent)
variable.
Happens a lot (mainly because anything can be dichotomized).
Some variable
An Introduction to Event History Analysis
Oxford Spring School
June 18-20, 2007
Day Three: Diagnostics, Extensions, and Other Miscellanea
Data Redux: Supreme Court Vacancies, 1789-1992
. stset service
POLS 8501
Advanced Quantitative Methods II
Spring 2014
3022 Baldwin
T 3.30-6.15
Instructor: Ryan Bakker
E-mail: rbakker [at] uga [dot] edu
Office: 416 Baldwin Hall
Office Hours: M W 1-3 or by appt.
De
An Introduction to Event History Analysis
Oxford Spring School
June 18-20, 2007
Day Two: Regression Models for Survival Data
Figure 1: Various Functions of an Exponential Model with = 0.02
1
Figure 2:
An Introduction to Event History Analysis
Oxford Spring School
June 18-20, 2007
Day One: Exploring Survival Data
Survival Analysis
Survival analysis is also known as event history analysis (sociology)
An Introduction to Event History Analysis
Oxford Spring School
June 18-20, 2007
Day One: Exploring Survival Data
Single-Record Data
The (completely made-up) example data:
. list id X T c
1.
2.
3.
4.
5
POLS 7050
Spring 2009
April 23, 2009
Pulling it All Together: Generalized Linear Models
Throughout the course, weve considered the various models weve discussed discretely. In
fact, however, most of t
POLS 7050
Spring 2008
April 9, 2007
Models for Event Count Data, II
Heterogeneity, Contagion, and OverdispersionOh My!
Event count models are similar to some of our earlier models (e.g. logits and pro
Advanced MLE
Introduction to Longitudinal Data
July 24, 2007
Introduction
As the course description suggests, this is a course on the analysis of longitudinal data.
Despite the apparent obviousness of
POLI 8501
Binary Logit & Probit, II
The topic du jour is interpretation of binary response models (i.e. logit and probit
estimates).
Yes, their interpretation is harder (more involved, acutally) than
POLI 7050
Spring 2008
March 5, 2008
Unordered Response Models II
Introduction
Today well talk about interpreting MNL and CL models. Well start with general issues of model
t, and then get to variable
POLS 7050
Spring 2008
April 9, 2007
Models for Event Count Data, I
An Introduction to Count Data
Event count models are models where the dependent variable is a count of events. So, were
considering a
POLI 8501
Models for Ordinal Responses I
It is often the case that we want to model variables that take the form of a small number of
discrete, ordered categories. WE can think of two types . . .
Gro
POLI 7050
Spring 2008
February 27, 2008
Unordered Response Models I
Introduction
For the next couple weeks well be talking about unordered, polychotomous dependent variables. Examples include:
Voter
POLI 8501
Binary Response Models, III
Heterscedasticity and Binary Response Models
Consider a basic latent-variable probit model with N observations and k independent variables Xi :
Yi = Xi + ui
(1)
w
POLI 8501
Models for Ordinal Responses II
Introduction
There are several possible ways of interpreting these models, all of which have similarities
to binary logit/probit:
Coecient and standard error
POLI 8501
Introduction to Maximum Likelihood Estimation
Maximum Likelihood
Intuition
Consider a model that looks like this:
Yi N (, 2 )
So:
E(Y ) =
V ar(Y ) = 2
Suppose you have some data on Y , and
An Introduction to Event History Analysis
Oxford Spring School
June 18-20, 2007
Day Two: Regression Models for Survival Data
Parametric Models
Well spend the morning introducing regression-like models