The course provides an elementary but
comprehensive look to the practice of modern econometrics. The main topics
covered include heteroscedasticity, consequences
of heteroscedasticity, detecting heteroscedasticity, Breusch-Pagan LM Test,
Harvey-Godfrey LM Test, Goldfeld-Quandt Test, White Test, resolving
heteroscedasticity, Generalized or Weighted Least Squares, White’s
heteroscedasticity consistent variances and standard errors, autocorrelation, consequences of
autocorrelation, detecting autocorrelation, graphical method, Durbin-Watson
Test, Breusch-Godfrey LM test for serial correlation, Durbin’s h test, resolving
autocorrelation, Generalized Least Squares (GLS), Estimated GLS (EGLS), Newey-West
method, common-factor test, apparent autocorrelation, autoregressive conditional heteroscedasticity
(ARCH) model, testing for ARCH effects, estimation of ARCH models, generalized
autoregressive conditional heteroscedasticity (GARCH) model, estimation of
GARCH Models, estimating dynamic models, adjustment lags, problems in the
estimation of dynamic models, formation of Expectations: adaptive expectations,
rational expectations, consequences and detection of errors of specification,
data mining, alternative approaches to selecting the best model, selecting
models: some important criteria, general to specific modelling, stationarity
and non-stationarity, unit roots and spurious regressions, testing for unit
roots, Dickey-Fuller test, Augmented Dickey-Fuller test, Phillips-Perron test, cointegration,
cointegration and error-correction model (ECM), cointegration in single
equations (The Engle-Granger approach, drawbacks of Engle Granger approach), vector
autoregressive (VAR) models, causality tests (Granger causality test, Sims
causality test), cointegration in multiple equations, Johansen approach, steps
of Johansen approach, identification in standard and cointegrated systems
(order condition, rank condition), traditional panel data models, fixed effects
model, random effects model, Hausman test, non-stationary panels, panel unit
root tests (Levin and Lin test; Im, Pesaran and Shin test, Maddala and Wu
test), panel cointegration tests (Kao test, McCoskey test, Pedroni tests,
Larsson test), simultaneous equation models, simultaneity, identification
problem, conditions of identification, estimation of an exactly identified
equation: indirect least squares (ILS), and estimation of an over-identified
equation: two-stage least squares (TSLS).