Class 07

Which key assumption is necessary for the RE model to provide consistent estimates?

• Homoscedasticity

• Linear functional form of the model

• Independence between the individual-specific effect and the regressors

What is the main advantage of using a Fixed Effects (FE) model?

• It allows for time-invariant unobserved heterogeneity.

• It allows for time-variant unobserved heterogeneity.

• It provides more efficient estimates than OLS.

• It can estimate coefficients for time-invariant variables.

What happens to time-invariant variables in the Fixed Effects (FE) model?

• They are included in the estimation.

• Their coefficients are estimated with higher efficiency.

• They are dropped from the estimation.

• They are treated as random noise.

In a Random Effects (RE) model, the individual-specific effect is

• Assumed to be uncorrelated with regressors.

• Included as part of the error term but without assumptions.

• Treated as fixed and correlated with regressors.

When would you prefer a Fixed Effects (FE) model over a Random Effects (RE) model?

• When regressors are correlated with the individual-specific effects.

• When time-invariant variables are important in the model.

• When the dataset is very small.

When should you prefer First Differences (FD) over Fixed Effects (FE) regarding autocorrelation in the error term?

• Always use FD

• Always use FE

• When the error term is weakly dependent

• When the error term follows a random walk