ECS4863 ||Exam Study Pack || 2026
ECS4863 – ADVANCED ECONOMETRICS
EXAM STUDY PACK 2026
SECTION A: CLASSICAL LINEAR REGRESSION AND OLS
1. Which assumption of the classical linear regression model requires that the
relationship between the dependent and independent variables is linear in
parameters?
A) Homoscedasticity
B) Linearity
C) Normality
D) No autocorrelation
Answer: B – Linearity
Explanation: Linearity means the model is linear in the unknown parameters,
allowing estimation using OLS .
2. Which estimator is considered BLUE (Best Linear Unbiased Estimator) under
the Gauss-Markov assumptions?
A) Maximum Likelihood Estimator
B) Instrumental Variable Estimator
C) Ordinary Least Squares (OLS) Estimator
D) Generalized Method of Moments Estimator
Answer: C – Ordinary Least Squares (OLS) Estimator
Explanation: OLS is BLUE when the Gauss-Markov assumptions hold .
,3. The primary objective of econometrics is to:
A) Eliminate uncertainty in economics
B) Develop accounting standards
C) Quantify and test economic relationships using statistical methods
D) Replace economic theory
Answer: C – Quantify and test economic relationships using statistical methods
Explanation: Econometrics combines economic theory, mathematics, and statistics
to estimate and test relationships .
4. Which measure indicates the proportion of variation in the dependent
variable explained by the regression model?
A) F-statistic
B) R-squared
C) Durbin-Watson statistic
D) Standard error of regression
Answer: B – R-squared
Explanation: R² measures the explanatory power of the regression model .
5. Multicollinearity primarily affects:
A) Unbiasedness of OLS estimates
B) Precision of coefficient estimates
C) Normality of residuals
D) Linearity of parameters
Answer: B – Precision of coefficient estimates
Explanation: Multicollinearity inflates standard errors, reducing statistical
precision .
6. Which test is commonly used to detect heteroskedasticity?
A) Durbin-Watson Test
,B) Breusch-Pagan Test
C) Jarque-Bera Test
D) Augmented Dickey-Fuller Test
Answer: B – Breusch-Pagan Test
Explanation: The Breusch-Pagan test evaluates whether error variance is constant
.
7. A dummy variable typically represents:
A) Continuous observations
B) Time-series observations
C) Qualitative or categorical information
D) Lagged variables
Answer: C – Qualitative or categorical information
Explanation: Dummy variables encode categorical characteristics using 0 and 1 .
8. Which assumption requires the expected value of the error term to be zero?
A) Homoscedasticity
B) Zero Conditional Mean
C) Normality
D) No autocorrelation
Answer: B – Zero Conditional Mean
Explanation: This assumption ensures unbiased parameter estimation .
9. Autocorrelation is most commonly associated with:
A) Cross-sectional data
B) Time-series data
C) Panel data only
D) Qualitative data
, Answer: B – Time-series data
Explanation: Serial dependence is primarily a feature of time-series observations .
10. Which statistic is commonly used to detect first-order autocorrelation?
A) Breusch-Pagan Statistic
B) Durbin-Watson Statistic
C) Jarque-Bera Statistic
D) ADF Statistic
Answer: B – Durbin-Watson Statistic
Explanation: The Durbin-Watson statistic measures first-order serial correlation .
11. Omitted variable bias occurs when:
A) A relevant variable is included in the model
B) A relevant variable is excluded and is correlated with included variables
C) The sample size is too large
D) The dependent variable is stationary
Answer: B – A relevant variable is excluded and is correlated with included
variables
Explanation: OVB occurs when an omitted variable both affects the dependent
variable and is correlated with included independent variables .
12. Positive omitted variable bias occurs when the omitted variable is:
A) Negatively correlated with both the dependent and included variable
B) Positively correlated with both the dependent and included variable
C) Uncorrelated with the dependent variable
D) Uncorrelated with the included variable
Answer: B – Positively correlated with both the dependent and included variable
Explanation: Positive bias inflates the estimated coefficient .
ECS4863 – ADVANCED ECONOMETRICS
EXAM STUDY PACK 2026
SECTION A: CLASSICAL LINEAR REGRESSION AND OLS
1. Which assumption of the classical linear regression model requires that the
relationship between the dependent and independent variables is linear in
parameters?
A) Homoscedasticity
B) Linearity
C) Normality
D) No autocorrelation
Answer: B – Linearity
Explanation: Linearity means the model is linear in the unknown parameters,
allowing estimation using OLS .
2. Which estimator is considered BLUE (Best Linear Unbiased Estimator) under
the Gauss-Markov assumptions?
A) Maximum Likelihood Estimator
B) Instrumental Variable Estimator
C) Ordinary Least Squares (OLS) Estimator
D) Generalized Method of Moments Estimator
Answer: C – Ordinary Least Squares (OLS) Estimator
Explanation: OLS is BLUE when the Gauss-Markov assumptions hold .
,3. The primary objective of econometrics is to:
A) Eliminate uncertainty in economics
B) Develop accounting standards
C) Quantify and test economic relationships using statistical methods
D) Replace economic theory
Answer: C – Quantify and test economic relationships using statistical methods
Explanation: Econometrics combines economic theory, mathematics, and statistics
to estimate and test relationships .
4. Which measure indicates the proportion of variation in the dependent
variable explained by the regression model?
A) F-statistic
B) R-squared
C) Durbin-Watson statistic
D) Standard error of regression
Answer: B – R-squared
Explanation: R² measures the explanatory power of the regression model .
5. Multicollinearity primarily affects:
A) Unbiasedness of OLS estimates
B) Precision of coefficient estimates
C) Normality of residuals
D) Linearity of parameters
Answer: B – Precision of coefficient estimates
Explanation: Multicollinearity inflates standard errors, reducing statistical
precision .
6. Which test is commonly used to detect heteroskedasticity?
A) Durbin-Watson Test
,B) Breusch-Pagan Test
C) Jarque-Bera Test
D) Augmented Dickey-Fuller Test
Answer: B – Breusch-Pagan Test
Explanation: The Breusch-Pagan test evaluates whether error variance is constant
.
7. A dummy variable typically represents:
A) Continuous observations
B) Time-series observations
C) Qualitative or categorical information
D) Lagged variables
Answer: C – Qualitative or categorical information
Explanation: Dummy variables encode categorical characteristics using 0 and 1 .
8. Which assumption requires the expected value of the error term to be zero?
A) Homoscedasticity
B) Zero Conditional Mean
C) Normality
D) No autocorrelation
Answer: B – Zero Conditional Mean
Explanation: This assumption ensures unbiased parameter estimation .
9. Autocorrelation is most commonly associated with:
A) Cross-sectional data
B) Time-series data
C) Panel data only
D) Qualitative data
, Answer: B – Time-series data
Explanation: Serial dependence is primarily a feature of time-series observations .
10. Which statistic is commonly used to detect first-order autocorrelation?
A) Breusch-Pagan Statistic
B) Durbin-Watson Statistic
C) Jarque-Bera Statistic
D) ADF Statistic
Answer: B – Durbin-Watson Statistic
Explanation: The Durbin-Watson statistic measures first-order serial correlation .
11. Omitted variable bias occurs when:
A) A relevant variable is included in the model
B) A relevant variable is excluded and is correlated with included variables
C) The sample size is too large
D) The dependent variable is stationary
Answer: B – A relevant variable is excluded and is correlated with included
variables
Explanation: OVB occurs when an omitted variable both affects the dependent
variable and is correlated with included independent variables .
12. Positive omitted variable bias occurs when the omitted variable is:
A) Negatively correlated with both the dependent and included variable
B) Positively correlated with both the dependent and included variable
C) Uncorrelated with the dependent variable
D) Uncorrelated with the included variable
Answer: B – Positively correlated with both the dependent and included variable
Explanation: Positive bias inflates the estimated coefficient .
