These questions cover all 12 chapters and are designed to test understanding of key concepts, interpretation, and application.
Chapter 1: Introduction to Econometrics and MORISTAT
1.1 What is econometrics? Explain the five steps of the econometric workflow.
See section 1.1.1
1.2 Explain the difference between interactive script mode and batch script mode in MORISTAT. When would you use each?
See section 1.2.2
1.3 Interpret the following MORISTAT regression output: Intercept = 10.2345, Advertising = 2.3456, Price = −1.2345, R² = 0.8924.
See section 1.3
1.4 What are the key features of MORISTAT that make it suitable for teaching econometrics?
See section 1.2.1
Chapter 2: Review of Statistics and Probability
2.1 Define the expected value and variance of a random variable. Provide formulas for both discrete and continuous cases.
See section 2.1.1
2.2 What is the difference between covariance and correlation? Explain the properties of correlation.
See section 2.1.2
2.3 A p-value of 0.03 is reported for a coefficient. At α = 0.05, do you reject or fail to reject H₀? Explain your reasoning.
See section 2.4
2.4 Explain the five steps of hypothesis testing. Provide an example using a regression coefficient.
See section 2.4.1
2.5 What is the Central Limit Theorem and why is it important for econometrics?
See section 2.3
Chapter 3: Simple Linear Regression
3.1 Write down the simple linear regression model and define each term. Explain the interpretation of β₀ and β₁.
See section 3.1
3.2 Derive the OLS estimators for β₀ and β₁ in simple linear regression.
See section 3.2
3.3 What is R-squared? How is it calculated, and what does it measure? Why is it not a measure of causality?
See section 3.5
3.4 State the Gauss-Markov Theorem. What are the assumptions required for OLS to be BLUE?
See section 3.8
3.5 Interpret the following MORISTAT output for a simple regression: Advertising coefficient = 2.3456, Std. Error = 0.4567, t-stat = 5.14, p-value = 0.0002.
See section 3.4
Chapter 4: Multiple Regression Analysis (Estimation)
4.1 Write down the multiple linear regression model with k independent variables. Explain the ceteris paribus interpretation of each coefficient.
See section 4.1
4.2 Explain the "partialling out" interpretation of multiple regression coefficients.
See section 4.2.2
4.3 What is perfect collinearity? Why is it a problem for OLS estimation?
See section 4.3
4.4 What is the difference between R-squared and Adjusted R-squared? When would you use Adjusted R-squared?
See section 4.5
4.5 Explain the concept of goodness of fit in multiple regression. What statistics are used to measure it?
See section 4.5
Chapter 5: Multiple Regression Analysis (Inference)
5.1 Explain the t-test for testing H₀: βⱼ = 0 vs H₁: βⱼ ≠ 0. Provide the test statistic formula and state the decision rule.
See section 5.1
5.2 What is the difference between a one-sided and a two-sided alternative hypothesis? Give examples of each.
See section 5.1.2
5.3 Interpret a 95% confidence interval for β₁: [0.4567, 0.7893].
See section 5.2
5.4 Explain the F-test for overall significance of a regression. How is the test statistic calculated?
See section 5.3
5.5 What are the consequences of using the wrong significance level (α) in hypothesis testing?
See section 5.1.3
Chapter 6: Specification Issues and Dummy Variables
6.1 What is omitted variable bias? How does it affect regression estimates?
See section 6.1
6.2 Explain the use of dummy variables in regression analysis. How are they interpreted when there are multiple categories?
See section 6.2
6.3 How do you test for the significance of a set of dummy variables?
See section 6.2.2
6.4 What is the RESET test and what does it diagnose?
See section 6.3
6.5 How do you interpret a coefficient on a dummy variable in a model with a log-transformed dependent variable?
See section 6.2.3
Chapter 7: Heteroskedasticity, Serial Correlation, and IV
7.1 What is heteroskedasticity? What are its consequences for OLS estimation?
See section 7.1
7.2 Explain the Breusch-Pagan test for heteroskedasticity. How is it conducted?
See section 7.1.2
7.3 What is serial correlation? When does it typically occur?
See section 7.2
7.4 What is the Durbin-Watson test and what does it test?
See section 7.2.1
7.5 Explain the instrumental variables (IV) approach. What are the conditions for a valid instrument?
See section 7.3
7.6 What is the Hausman-Wu test and why is it used?
See section 7.3.2
Chapter 8: Panel Data and Difference-in-Differences
8.1 What is panel data? What are the advantages of using panel data over cross-sectional data?
See section 8.1
8.2 Explain the Fixed Effects (FE) model. How does it control for unobserved heterogeneity?
See section 8.2
8.3 What is the Random Effects (RE) model? What is the key assumption that distinguishes it from FE?
See section 8.3
8.4 Explain the Hausman test for choosing between FE and RE.
See section 8.4
8.5 What is the Difference-in-Differences (DiD) estimator? What are the key assumptions?
See section 8.5
8.6 How do you test the parallel trends assumption in DiD?
See section 8.5.2
Chapter 9: Time Series Analysis and Forecasting
9.1 What is a time series? What are the key features of time series data?
See section 9.1
9.2 What is stationarity? Why is it important in time series analysis?
See section 9.2
9.3 Explain the concept of autocorrelation in time series.
See section 9.2.1
9.4 What is the ARIMA model? Explain the components p, d, and q.
See section 9.3
9.5 How do you evaluate a forecasting model?
See section 9.4
9.6 What is the difference between in-sample and out-of-sample forecasting?
See section 9.4.1
Chapter 10: Limited Dependent Variable Models (Logit/Probit)
10.1 What are limited dependent variable models? Give examples of when they are used.
See section 10.1
10.2 Explain the Logit model. What is the link function?
See section 10.2
10.3 What is the difference between Logit and Probit models?
See section 10.3
10.4 How do you interpret Logit coefficients?
See section 10.2.2
10.5 What are marginal effects and why are they useful in binary choice models?
See section 10.4
10.6 What is the pseudo R-squared? How is it different from the standard R-squared?
See section 10.5
Chapter 11: Machine Learning for Econometrics
11.1 What is the distinction between supervised and unsupervised learning?
See section 11.1
11.2 What is the bias-variance tradeoff? How does it relate to model complexity?
See section 11.2
11.3 What is cross-validation? Why is it important in machine learning?
See section 11.2.2
11.4 Explain Ridge regression. How does it differ from OLS?
See section 11.3
11.5 What is Lasso regression? How does it perform variable selection?
See section 11.4
11.6 How do you evaluate the performance of a classification model?
See section 11.5
Chapter 12: Epilogue — The Responsible Use of Statistics
12.1 What are the key ethical responsibilities of an econometrician?
See section 12.1
12.2 What is the difference between correlation and causation? Why is this distinction important?
See section 12.2
12.3 Why is replication important in empirical research?
See section 12.3
12.4 What are some common biases that can arise in econometric analysis?
See section 12.4
12.5 How does MORISTAT support the principles of open and reproducible research?
See section 12.5
Comprehensive Questions
C1. You are given data on income, education, and experience. Outline the steps you would take to estimate the returns to education using a multiple regression model. Include model specification, estimation, and hypothesis testing.
C2. A researcher is concerned that the relationship between education and earnings suffers from omitted variable bias. What approaches could they use to address this issue?
C3. You are interested in the effect of a policy intervention on economic growth. You have data on 100 countries over 30 years. Describe the econometric approach you would use and justify your choice.
C4. Compare and contrast the use of OLS, Logit, and Lasso regression. When would you use each?
C5. Explain how MORISTAT can be used as a teaching tool for econometrics. What features make it particularly suitable?