Econ 570 -- Fall 2006

Midterm 2 Answers

Time Series Econometrics

We had a bonus lecture due to a lack of questions about the old exams.

]]>

Review

MT 1 discussion

From 10/24/06:

Extensions of logit/probit

(The leading term in the denominator can either be the exponential function or 1.)

logit/probit

Calling this the "linear probability model" makes it sound better than it really is. The regression model is simply not the best option for categorical data.

The probit model is designed for this situation.

The logit model is another excellent choice.

The odds ratio

leads to the log odds ratio.

The likelihood ratio test is our bread and butter test for logit/probit models.

Term Paper Research

Term Papers & Misc

Simultaneity, Identification, & Two-Stage Least Squares

Heteroskedasticity

Somebody forgot his camera.

Autocorrelated Errors

Structural Change

BLS Income Data

The BLS income data is at the following links.

]]> You will need the data and the variable definitions. The do file
from today's class might also be helpful.

Elasticities

From the Q & A session:

]]>