Week 7 - Regression Diagnosis (II): Endogeneity Flashcards
external vs internal validity
External validity: Inferences generated in your model can be
extended from your population and settings of study to other
populations with different settings.
2 Internal validity: Causal claims inferred from your model are truly
valid for the population under study.
Threats to internal validity
Endogeneity
When do you observed omitted variable bias?
OVB Problem 1
OVB Problem 2
Summary of direction of bias in ommited variable equation.
When do you need to use Instrumental Variable?
IV with a single regressor and a single instrument
Getting the right terminology
An endogenous variable is one that is correlated with uˆ.
2 An exogenous variable is one that is uncorrelated with uˆ.
Conditions for a valid instrument
Estimation: 2SLS Stage 1
Estimation: 2SLS Stage 2
2SLS with a single endogenous regressor a m instruments
Checking assumption #1: Instrument relevance
The strength of Instruments: The First-stage F-statistics
The first stage regression (one X):
2 Regress X on Z1, .., Zm, W1, . . . , Wk .
3 Totally irrelevant instruments =⇒ all the coefficients on Z1, .., Zm are
zero.
4 The first-stage F-statistic tests the hypothesis that Z1, .., Zm do not
enter the first stage regression.
5 Weak instruments imply a small first stage F-statistic.
Where do valid instruments come from?
Method #1
1 Variables in another equation (factors that explain GDP/cap but are
not necessarily relate to Foreign Aid).
2 Method #2
1 Look for exogenous variation (Z) that is “ as if ” randomly assigned
(does not directly affect Y) but affects X.
