Question 751
A regression model with too many irrelevant variables is likely to:
A. Have inflated standard errors
B. Have perfect predictions
C. Have no residuals
D. Have a negative intercept
Check answer
Select an option first.
Correct answer: A — Have inflated standard errors
Explanation: Extra noise increases variance of estimates. B/C/D: Not consequences.
Question 752
A model with too few variables (omitting relevant ones) may suffer from:
A. Omitted variable bias
B. Perfect accuracy
C. Zero residuals
D. No intercept
Check answer
Select an option first.
Correct answer: A — Omitted variable bias
Explanation: Missing relevant predictors biases coefficients. B/C/D: Incorrect.
Question 753
A regression model with a very small sample size is likely to have:
A. Highly reliable estimates
B. Unstable and imprecise estimates
C. No residuals
D. Perfect multicollinearity
Check answer
Select an option first.
Correct answer: B — Unstable and imprecise estimates
Explanation: Small samples → high uncertainty. A/C/D: Not necessarily true.
Question 754
The standard error of a coefficient measures:
A. The average size of the residuals
B. The uncertainty around the estimated coefficient
C. The correlation between variables
D. The intercept’s magnitude
Check answer
Select an option first.
Correct answer: B — The uncertainty around the estimated coefficient
Explanation: Standard error quantifies estimation uncertainty. A/C/D: Not correct.
Question 755
A wide confidence interval for a coefficient suggests:
A. High precision
B. Low precision
C. Perfect fit
D. No residuals
Check answer
Select an option first.
Correct answer: B — Low precision
Explanation: Wide interval = high uncertainty. A/C/D: Incorrect.
Get the full CFA question bank — free
Drop your email and we'll send you fresh CFA practice questions, fully worked solutions and exam-deadline reminders. No spam — unsubscribe in one click.
Question 756
A model with a high (R^2) but low adjusted (R^2) likely has:
A. Too few variables
B. Too many irrelevant variables
C. No intercept
D. No residuals
Check answer
Select an option first.
Correct answer: B — Too many irrelevant variables
Explanation: Adjusted (R^2) penalizes unnecessary predictors. A/C/D: Not implied.
Question 757
A residual with a large absolute value indicates:
A. A good prediction
B. A poor prediction for that observation
C. A perfect fit
D. No error
Check answer
Select an option first.
Correct answer: B — A poor prediction for that observation
Explanation: Large residual = large prediction error. A/C/D: Incorrect.
Question 758
If residuals are centered around zero, this supports the assumption that:
A. Errors have mean zero
B. Errors have increasing variance
C. Errors are perfectly correlated
D. Errors are always positive
Check answer
Select an option first.
Correct answer: A — Errors have mean zero
Explanation: Mean-zero residuals support the assumption. B/C/D: Not implied.
Question 759
A model with a very high (R^2) (e.g., 0.99) in economic data may indicate:
A. A perfect model
B. Possible overfitting or nonstationarity
C. No issues
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Possible overfitting or nonstationarity
Explanation: Extremely high fit in noisy data is suspicious. A/C/D: Not necessarily true.
Question 760
A regression coefficient close to zero suggests:
A. The variable has little linear relationship with the dependent variable
B. The variable is irrelevant in all contexts
C. The model is invalid
D. The intercept must be removed
Check answer
Select an option first.
Correct answer: A — The variable has little linear relationship with the dependent variable
Explanation: Small coefficient → weak linear effect. B/C/D: Not implied.
Question 761
A model with a negative intercept is:
A. Automatically invalid
B. Acceptable depending on context
C. Impossible
D. A sign of multicollinearity
Check answer
Select an option first.
Correct answer: B — Acceptable depending on context
Explanation: Intercept sign depends on scaling and context. A/C/D: Incorrect.
Question 762
A regression model with a single extremely influential data point may produce:
A. Stable coefficients
B. Distorted coefficients
C. Zero residuals
D. Perfect predictions
Check answer
Select an option first.
Correct answer: B — Distorted coefficients
Explanation: Outliers can distort estimates. A/C/D: Not typical.
Question 763
A key assumption of regression is that errors are:
A. Correlated with independent variables
B. Uncorrelated with independent variables
C. Always positive
D. Always negative
Check answer
Select an option first.
Correct answer: B — Uncorrelated with independent variables
Explanation: Errors must be independent of predictors. A/C/D: Incorrect.
Question 764
If errors are correlated with an independent variable, this may indicate:
A. Perfect model specification
B. Omitted variable bias
C. No issues
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Omitted variable bias
Explanation: Omitted variables often cause correlation with errors. A/C/D: Incorrect.
Question 765
A model that uses a linear form when the true relationship is curved suffers from:
A. Correct specification
B. Incorrect functional form
C. Perfect fit
D. No residuals
Check answer
Select an option first.
Correct answer: B — Incorrect functional form
Explanation: Wrong functional form = misspecification. A/C/D: Incorrect.
Question 766
A scatterplot of residuals showing a curved pattern suggests:
A. Linearity holds
B. Linearity may be violated
C. No issues
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Linearity may be violated
Explanation: Curvature indicates non-linearity. A/C/D: Incorrect.
Question 767
A model with heteroskedastic errors may produce:
A. Unbiased coefficients but unreliable standard errors
B. Biased coefficients
C. Zero residuals
D. Perfect predictions
Check answer
Select an option first.
Correct answer: A — Unbiased coefficients but unreliable standard errors
Explanation: Coefficients remain unbiased; inference becomes unreliable. B/C/D: Incorrect.
Question 768
A model with serial correlation in residuals may produce:
A. Reliable standard errors
B. Underestimated standard errors
C. No issues
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Underestimated standard errors
Explanation: Serial correlation often understates uncertainty. A/C/D: Incorrect.
Question 769
A Durbin–Watson statistic near 2 suggests:
A. No serial correlation
B. Strong positive serial correlation
C. Strong negative serial correlation
D. Zero residuals
Check answer
Select an option first.
Correct answer: A — No serial correlation
Explanation: DW ≈ 2 → no autocorrelation. B/C/D: Incorrect.
Question 770
A variance inflation factor (VIF) is used to detect:
A. Heteroskedasticity
B. Multicollinearity
C. Serial correlation
D. Non-linearity
Check answer
Select an option first.
Correct answer: B — Multicollinearity
Explanation: VIF measures collinearity among predictors. A/C/D: Different diagnostics.
Question 771
A very high VIF (e.g., >10) suggests:
A. No multicollinearity
B. Strong multicollinearity
C. Zero residuals
D. Perfect fit
Check answer
Select an option first.
Correct answer: B — Strong multicollinearity
Explanation: High VIF = strong collinearity. A/C/D: Incorrect.
Question 772
A model with multicollinearity may still:
A. Predict well
B. Predict poorly
C. Have no residuals
D. Have zero variance
Check answer
Select an option first.
Correct answer: A — Predict well
Explanation: Prediction can remain strong even with collinearity. B/C/D: Not necessarily true.
Question 773
A model with omitted variables may show:
A. Stable coefficients
B. Biased coefficients
C. Zero residuals
D. Perfect predictions
Check answer
Select an option first.
Correct answer: B — Biased coefficients
Explanation: Omitted variables bias estimates. A/C/D: Incorrect.
Question 774
A model with residuals that increase over time suggests:
A. Homoskedasticity
B. Heteroskedasticity
C. No issues
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Heteroskedasticity
Explanation: Increasing variance → heteroskedasticity. A/C/D: Incorrect.
Question 775
A stationary time series has:
A. Constant mean and variance
B. Increasing variance
C. No mean
D. No randomness
Check answer
Select an option first.
Correct answer: A — Constant mean and variance
Explanation: Stationarity requires constant mean/variance. B/C/D: Incorrect.
Question 776
A nonstationary series often shows:
A. Constant mean
B. Trends or changing variance
C. No randomness
D. Zero residuals
Check answer
Select an option first.
Correct answer: B — Trends or changing variance
Explanation: Nonstationary series drift or change over time. A/C/D: Incorrect.
Question 777
A time-series model that uses lagged values of the dependent variable is:
A. MA model
B. AR model
C. Dummy model
D. Logistic model
Check answer
Select an option first.
Correct answer: B — AR model
Explanation: AR uses lagged dependent values. A/C/D: Incorrect.
Question 778
A moving-average model uses:
A. Lagged dependent values
B. Lagged error terms
C. Future values
D. Dummy variables
Check answer
Select an option first.
Correct answer: B — Lagged error terms
Explanation: MA uses past shocks. A/C/D: Incorrect.
Question 779
A random walk has:
A. Strong mean reversion
B. No mean reversion
C. Constant variance
D. Zero randomness
Check answer
Select an option first.
Correct answer: B — No mean reversion
Explanation: Random walks do not revert to a mean. A/C/D: Incorrect.
Question 780
A unit root indicates:
A. Stationarity
B. Nonstationarity
C. Zero variance
D. Perfect fit
Check answer
Select an option first.
Correct answer: B — Nonstationarity
Explanation: Unit root → nonstationary. A/C/D: Incorrect.
More free CFA topics