« October 2005 | Main | December 2005 »
November 15, 2005
Review of Spring 2005 Midterm 2
Posted by bparke at 10:54 PM | Comments (0)
November 10, 2005
Serially Correlated Errors - III
Forecasting.
Posted by bparke at 10:49 PM | Comments (0)
Interesting Time Series Data
The Wall Street Journal "C" Section (Money and Investing) basically gives a daily discussion of the choice among stocks, bonds, and cash. The empirical research in this area often faces serially correlated errors. DATA.
Posted by bparke at 08:49 AM | Comments (0)
Term Paper
Posted by bparke at 08:36 AM | Comments (0)
November 08, 2005
Serially Correlated Errors - II
Serially correlated error would violate the assumption that the errors are independent of each other.
The Cochrane-Orcutt estimation procedure alternates between estimating rho and estimating beta until the process converges.
The Durbin-Watson statistic is very nearly equal to 2(1-rho), where rho is the correlation between the current and lagged error residual.
A lagged dependent variable can induce dynamics very similar to those created by serially correlated errors. Distinguishing between the models is very difficult in practice.
Posted by bparke at 10:02 PM | Comments (0)
November 03, 2005
More Logs
The log-log model has a constant elasticity.
Log scales are used for sound levels (decibels) and hurricane strengths. They also make sense for income.
Posted by bparke at 09:01 PM | Comments (0)
Monte Carlo Simulation
We used the model with serially correlated errors as an example of simulation exercise.
Monte Carlo simulation allows us to tackle problems that are too difficult for analytic solution. It also helps us determine whether a proposition seems to be true so that we do not spend years trying to prove something that is not true.
How to construct normally distributed random numbers from uniformly distributed random numbers:
Posted by bparke at 08:56 PM | Comments (0)
November 01, 2005
Chapter 9 Exercises
We had a wide-ranging discussion of the homework.
Posted by bparke at 08:46 PM | Comments (0)