Advanced Panel Data Ch.15 Test Bank Docx

Real Stats – Chapter 15

True or False Questions:

1. True or False: Autocorrelation causes bias in a model with fixed effects.
2. True or False: Autocorrelation causes bias in a model without fixed effects.
3. True or False: A biased coefficient is always worse than an unbiased coefficient.
4. True or False: Random effects models are biased if the random effects error term is correlated with X.
5. True or False: Random effects models are appropriate only when their coefficients are similar to coefficients produced by fixed effects models.

Multiple Choice Questions:

1. Which of the following is a proper procedure for testing for autocorrelation in a panel data model without a lagged dependent variable?
   1. Run a model that does not address autocorrelation and test for autocorrelated errors based on the residuals from that model.
   2. Run a -transformed model and test for autocorrelated errors based on the residuals from that model.
   3. Assess whether there is a different intercept for each unit.
   4. Run a probit or logit model and test for autocorrelated errors based on the residuals from that model.
2. Under what conditions is the bias caused by autocorrelaed errors in a fixed effect model the most severe?
   1. We have many observations for each unit.
   2. We have few observations for each unit.
   3. We have many observations for each time period.
   4. We have few observations for each time period.
3. Suppose that we have the following results from a model with lagged dependent variable

Y_it = Y_i,t-1 + X_1it

Which of the following is most accurate?

• 1. The short-term effect of a one unit change in X will be a 2.9 unit change in Y.
  2. The long-term effect of a one unit change in X will be a 20 unit change in Y.
  3. The long-term effect of a one unit change in X will be a 4 unit change in Y.
  4. Y will keep getting smaller and smaller over time.

1. Suppose that we are comparing results in the following two models

Model A: Y_it = Y_i,t-1 + X_1it

Model A: Y_it = Y_i,t-1 + X_1it

Which of the following is true?

• 1. The long-term effect of a one unit change in X will be 10 times bigger in Model A than in Model B.
  2. The long-term effect of a one unit change in X will be 10 percent bigger in Model A than in Model B.
  3. The short-term effect of a one unit change in X will be bigger in Model A than in Model B.
  4. The short-term effect of a one unit change in X will be smaller in Model A than in Model B.

1. What usually happens to autocorrelation when we add a lagged dependent variable to the model?
   1. Autocorrelation gets bigger.
   2. Autocorrelation gets smaller.
   3. Autocorrelation stays the same.
   4. We cannot test for autocorrelation when a lagged dependent variable is included in the model.
2. What usually happens to autocorrelation when we add a lagged dependent variable to the model?
   1. Autocorrelation gets bigger.
   2. Autocorrelation gets smaller.
   3. Autocorrelation stays the same.
   4. We cannot test for autocorrelation when a lagged dependent variable is included in the model.
3. OLS coefficients are biased when there is a lagged dependent variable and fixed effects in the same model. Which of the following statements is most accurate about using OLS when there is a lagged dependent variable and fixed effects in the same model?
   1. We should never use OLS in this situation.
   2. Instrumental variables are easier to use OLS in this situation.
   3. OLS can produce coefficient estimates that are biased, but still on average closer to the true values in this situation.
   4. OLS performs worse in this situation when we have many observations for each unit.
4. What factor is associated with lower bias in a model with a lagged dependent variable and fixed effects?
   1. Large number of observations for each period.
   2. Large number of time periods.
   3. A large coefficient on the lagged dependent variable.
   4. Using multiple lags (e.g., the dependent variable from the previous period and the period before that).
5. Describe the potential trade-off between bias and accuracy using an example of an estimator that is biased, but accurate and one that is unbiased, but inaccurate.

For LDV models with fixed effects models, most scholars view the OLS approach as being biased, but reasonably accurate when we have a good number of periods (at least 10, and preferably more than 20).

1. Describe when it is appropriate to use random effects models.
2. Explain why a model with a lagged dependent variable and fixed effects will be biased.
3. Explain why it is important to test for autocorrelation when using a model with a lagged dependent variable and no fixed effects.