MEcon Master of Economics Econometrics and Research Methods 2 — Questions and Answers
Question 1: A Type I error in hypothesis testing is:
- Failing to reject a false null hypothesis
- Rejecting a true null hypothesis (Correct answer)
- Accepting the alternative when both are true
- Using the wrong test statistic
Correct answer: Rejecting a true null hypothesis
A Type I error (false positive) occurs when the null hypothesis is actually true but the test leads you to reject it; its probability is controlled by the significance level α.
Question 2: The p-value in hypothesis testing represents:
- The probability that the null hypothesis is true
- The probability of observing results at least as extreme as the data if the null hypothesis is true (Correct answer)
- The power of the test
- The effect size of the treatment
Correct answer: The probability of observing results at least as extreme as the data if the null hypothesis is true
The p-value is a conditional probability — how likely the observed (or more extreme) data would be if H₀ were true — not the probability that H₀ itself is true.
Question 3: Multicollinearity in a regression refers to:
- Correlation between the dependent variable and the error
- High correlation among two or more independent variables (Correct answer)
- Autocorrelation of residuals over time
- Non-linearity in the relationship between X and Y
Correct answer: High correlation among two or more independent variables
Multicollinearity makes it difficult to isolate the individual effect of each correlated regressor, inflating standard errors and making coefficient estimates unstable.
Question 4: A unit root in a time series indicates that the series is:
- Mean-reverting and stationary
- Non-stationary with shocks having permanent effects (Correct answer)
- Seasonally adjusted
- Normally distributed
Correct answer: Non-stationary with shocks having permanent effects
A unit root (random walk component) means the series does not return to a fixed mean — shocks permanently alter its level, making standard regression inference invalid.
Question 5: The R-squared statistic in regression measures:
- The causal effect of X on Y
- The proportion of variance in Y explained by the model (Correct answer)
- Whether the regression coefficients are statistically significant
- The correlation between residuals and fitted values
Correct answer: The proportion of variance in Y explained by the model
R² indicates goodness of fit — the share of total variation in the dependent variable accounted for by the regressors — but high R² does not imply causality.
Question 6: In panel data analysis, fixed effects estimation controls for:
- Time-varying unobservable heterogeneity
- Time-invariant unobservable unit-specific heterogeneity (Correct answer)
- Heteroskedasticity in cross-sectional data
- Serial correlation in purely time-series data
Correct answer: Time-invariant unobservable unit-specific heterogeneity
Fixed effects (within estimator) removes all time-invariant unit characteristics — observed or not — by demeaning or using unit dummies, eliminating a source of omitted variable bias.
MEcon Master of Economics Econometrics and Research Methods 2
A Type I error in hypothesis testing is: