|
Spring
2001
Instructors:
Dr.
Pavel Katyshev
pkatish@nes.ru, room 908,
Dr. Anatoly
Peresetsky, perstsky@cemi.rssi.ru, room 908,
TAs:
Sergei Golovan sgolovan@nes.ru,
Dmitry Shapiro dmshapir@nes.ru.
General
Information. This course is a continuation of the course "Econometrics-1".
This course meets Mondays 10:00, Wednesdays 12:00, Sections meets Wednesdays
14:30. Office hours are Mondays, Wednesdays room 908, from 17:45-18:45.
If you have any questions come and ask.
Texts.
The
main textbook for this course is:
П.К.Катышев, Я.Р.Магнус,
А.А.Пересецкий. Эконометрика. Начальный курс. 3-е издание, Дело, Москва,
2000.
Supplementary
reading.
- С.А.Айвазян, В.С.Мхитарян,
Прикладная статистика и основы эконометрики, ЮНИТИ, Москва, 1998.
- R.S. Pindyck
& D.L. Rubinfeld, Econometric Models and Economic Forecasts,
3rd edition, McGraw Hill, 1991.
- J.Johnston,
J.DiNardo, Econometrics Methods, 4th edition, McGraw-Hill,
1997.
- J.D.Hamilton,
Time Series Analysis, Princeton University Press, 1994.
- П.К.Катышев, А.А.Пересецкий,
Сборник задач к начальному курсу эконометрики. Дело, Москва, 1999.
Project.
A project will be suggested to students. Students can work on projects
in teams (no more than four students in a team). Project reports are
due April 25.
Homework
and Exams. Homework will be assigned and will be due each Wednesday.
Homework will be graded. There will be only final exam. Tentative dates
are:
Project
deadline April 25, 2001
Final
exam April 28, 2001 (the date could be revised).
Policy
on examination. A4 - format paper with your own notes. Xerox copies,
printed outputs, books are not allowed.
Grading.
The homework, the project, final exam will have the following weights:
Homework 0.15
Project 0.25
Final exam 0.60
The
final grade will be based on the final score, which is a weighted average
of the homework, the project and the final exam.
Bonus
points. Students, who will be firsts to find misprints in the main
textbook, may gain extra points to their scores.
COURSE
OUTLINE
I.
MAXIMUM LIKELIHOOD ESTIMATION
Three
lectures
Maximum
likelihood estimation (MLE): examples and formal treatment.
MLE of the multivariate normal distribution.
Properties of ML estimators.
MLE for linear regression model.
Three general criteria for testing hypothesis: likelihood ratio test,
Wald test, Lagrange multipliers test.
Criteria for testing hypotheses in the linear model.
Likelihood ratio, Wald, Lagrange multipliers tests in classical regression
model for testing general linear restriction.
II.
MODELS WITH LIMITED DEPENDENT VARIABLES
Four
lectures
- Discrete dependent
variables: qualitative (nominal), ranking, counted dependent variables.
- Binary choice
models. Linear probability model. Probit and Logit models. Interpretation
of the coefficients in binary choice models. Maximum likelihood
estimates in Probit and Logit models.
- Specification
errors in binary choice models. Multi-choice models.
- Models with
truncated and censored dependent variables. Tobit model. Biasedness
and inconsistency of OLS estimates. ML estimates.
- Duration models.
MODELS
WITH LAGGED VARIABLES AND TIME SERIES
Seven
lectures
1.
Models with lagged variables.
Distributed lags models.
Estimation of the distributed lags models. Polynomial lags (method Almon).
Geometric lags (Koyck model).
2.
Dynamic models.
Autoregressive model with autocorrelated errors. Estimation. Tests for
error autocorrelation (Durbin and Lagrange multiplier tests).
Examples of the models with lagged variables. (Partial adjustment model,
adaptive expectation model, error correction model). Granger causality
test.
3.
Unit roots and cointegration.
Stationarity. Random walk. AR(p) process. Unit roots. Dickey-Fuller
statistic. Augmented Dickey-Fuller test. Spurious regression. Cointegration.
Engel and Granger approach. MakKinnon statistic. Cointegration vector.
Long-run dynamic equilibrium.
4.
Box-Jenkins model (ARIMA).
Trend, seasonality, differenciing. Tests for stationarity. ACF and PACF.
Yule-Walker equations. MA models. Invertibility. Properties of ARMA
models.
Box-Jenkins methodology. Model identification. Estimation and diagnostic
checking. Ljung-Box test. Akaike and Schwarz criteria. Forecasting with
an ARIMA models. Seasonal ARIMA models.
5.
GARCH models.
ARCH, GARCH, ARCH-M, E-GARCH models. Lagrange multiplier test for ARCH.
Estimation methods.
|
