Question 1141
Bootstrap resampling draws samples:
A. Without replacement
B. With replacement
C. Only from normally distributed data
D. Only from lognormally distributed data
Check answer
Select an option first.
Correct answer: B — With replacement
Explanation: ✔ Why correct: Bootstrap requires sampling with replacement so observations can appear multiple times.
Question 1142
Monte Carlo simulation is especially useful when:
A. Analytical solutions are simple
B. The distribution of outcomes is complex
C. The sample size is extremely small
D. The variables are deterministic
Check answer
Select an option first.
Correct answer: B — The distribution of outcomes is complex
Explanation: ✔ Why correct: Monte Carlo handles complex, multi-factor, or nonlinear distributions.
Question 1143
A key difference between Monte Carlo and bootstrap is that Monte Carlo:
A. Uses the sample as the population
B. Requires assumed probability distributions
C. Cannot estimate risk
D. Does not use random numbers
Check answer
Select an option first.
Correct answer: B — Requires assumed probability distributions
Explanation: ✔ Why correct: Monte Carlo simulation draws from assumed distributions, not the sample.
Question 1144
Bootstrap resampling is particularly useful for estimating:
A. Expected returns only
B. Standard errors and sampling distributions
C. Maximum drawdowns
D. Value at Risk (VaR) only
Check answer
Select an option first.
Correct answer: B — Standard errors and sampling distributions
Explanation: ✔ Why correct: Bootstrap is widely used to estimate standard errors and sampling distributions when analytical formulas are unavailable.
Question 1145
Probability sampling is preferred because it:
A. Guarantees zero sampling error
B. Gives every population member a known chance of selection
C. Is always cheaper
D. Eliminates the need for a sampling plan
Check answer
Select an option first.
Correct answer: B — Gives every population member a known chance of selection
Explanation: ✔ Why correct: Probability sampling ensures each member has a known, non-zero chance of being selected, making the sample more representative.
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 1146
A simple random sample is one in which:
A. Only the largest observations are selected
B. Each population element has an equal probability of selection
C. Observations are selected based on researcher judgment
D. Observations are selected based on convenience
Check answer
Select an option first.
Correct answer: B — Each population element has an equal probability of selection
Explanation: ✔ Why correct: Equal probability of selection defines simple random sampling.
Question 1147
Simple random sampling is most appropriate when:
A. The population is homogeneous
B. The population is highly heterogeneous
C. The researcher wants to oversample certain groups
D. The population cannot be identified
Check answer
Select an option first.
Correct answer: A — The population is homogeneous
Explanation: ✔ Why correct: Simple random sampling works best when population characteristics are broadly similar.
Question 1148
Systematic sampling selects:
A. Every kth member of the population
B. Only members chosen by judgment
C. Only members chosen by convenience
D. Only members from each subgroup
Check answer
Select an option first.
Correct answer: A — Every kth member of the population
Explanation: ✔ Why correct: Systematic sampling selects every kth element after a random start.
Question 1149
Sampling error is:
A. Errors caused by incorrect data recording
B. The difference between a sample statistic and the population parameter
C. Errors caused by biased sampling
D. Errors caused by non-random sampling only
Check answer
Select an option first.
Correct answer: B — The difference between a sample statistic and the population parameter
Explanation: ✔ Why correct: Sampling error arises because a sample is only a subset of the population.
Question 1150
A sample statistic is a random variable because:
A. It changes depending on which sample is drawn
B. It is always equal to the population parameter
C. It is fixed and does not vary
D. It is chosen by the researcher
Check answer
Select an option first.
Correct answer: A — It changes depending on which sample is drawn
Explanation: ✔ Why correct: Different samples produce different sample statistics.
Question 1151
The sampling distribution of a statistic refers to:
A. The distribution of the population
B. The distribution of all possible values of the statistic across repeated samples
C. The distribution of measurement errors
D. The distribution of convenience samples
Check answer
Select an option first.
Correct answer: B — The distribution of all possible values of the statistic across repeated samples
Explanation: ✔ Why correct: Sampling distribution = distribution of a statistic across repeated samples.
Question 1152
Stratified random sampling is preferred when:
A. The population is homogeneous
B. Subgroups of interest must be represented
C. The researcher wants to minimize cost only
D. The population cannot be divided into strata
Check answer
Select an option first.
Correct answer: B — Subgroups of interest must be represented
Explanation: ✔ Why correct: Stratified sampling ensures proportional representation of key subgroups.
Question 1153
A key advantage of stratified random sampling is that it:
A. Produces higher variance
B. Guarantees representation of important subgroups
C. Requires no knowledge of population structure
D. Is always cheaper
Check answer
Select an option first.
Correct answer: B — Guarantees representation of important subgroups
Explanation: ✔ Why correct: Stratified sampling ensures each subgroup is included proportionally.
Question 1154
Cluster sampling differs from stratified sampling because cluster sampling:
A. Samples from every subgroup
B. Assumes each cluster is a mini-population
C. Requires homogeneous clusters
D. Always produces lower variance
Check answer
Select an option first.
Correct answer: B — Assumes each cluster is a mini-population
Explanation: ✔ Why correct: Clusters are assumed to be representative of the entire population.
Question 1155
In one-stage cluster sampling:
A. Only a subsample of each selected cluster is used
B. All elements in selected clusters are sampled
C. Only one cluster is sampled
D. Clusters must be identical in size
Check answer
Select an option first.
Correct answer: B — All elements in selected clusters are sampled
Explanation: ✔ Why correct: One-stage cluster sampling includes all members of selected clusters.
Question 1156
In two-stage cluster sampling:
A. All elements in selected clusters are sampled
B. A subsample is drawn from each selected cluster
C. Only one cluster is selected
D. Clusters must be homogeneous
Check answer
Select an option first.
Correct answer: B — A subsample is drawn from each selected cluster
Explanation: ✔ Why correct: Two-stage sampling selects clusters, then samples within them.
Question 1157
Convenience sampling is best described as:
A. A probability-based method
B. A method based on ease of access
C. A method that guarantees representativeness
D. A method that eliminates sampling error
Check answer
Select an option first.
Correct answer: B — A method based on ease of access
Explanation: ✔ Why correct: Convenience sampling selects elements based on accessibility.
Question 1158
Judgmental sampling is appropriate when:
A. The researcher has expertise that improves sample selection
B. The population is homogeneous
C. Probability sampling is required
D. Sampling error must be minimized statistically
Check answer
Select an option first.
Correct answer: A — The researcher has expertise that improves sample selection
Explanation: ✔ Why correct: Judgmental sampling leverages expert insight to select meaningful observations.
Question 1159
A major drawback of judgmental sampling is that it:
A. Is too expensive
B. Introduces researcher bias
C. Requires too much data
D. Cannot be used in auditing
Check answer
Select an option first.
Correct answer: B — Introduces researcher bias
Explanation: ✔ Why correct: Judgmental sampling is vulnerable to subjective bias.
Question 1160
The central limit theorem states that the sampling distribution of the sample mean:
A. Is always normal
B. Approaches normality as sample size increases
C. Is always skewed
D. Has infinite variance
Check answer
Select an option first.
Correct answer: B — Approaches normality as sample size increases
Explanation: ✔ Why correct: CLT ensures approximate normality for large samples, regardless of population shape (with finite variance).
Question 1161
The central limit theorem applies when:
A. The population has infinite variance
B. The sample size is sufficiently large
C. The population is always normal
D. The sample is non-random
Check answer
Select an option first.
Correct answer: B — The sample size is sufficiently large
Explanation: ✔ Why correct: A sufficiently large random sample is the key requirement for CLT.
Question 1162
Bootstrap resampling draws samples:
A. Without replacement
B. With replacement
C. Only from normally distributed data
D. Only from stratified samples
Check answer
Select an option first.
Correct answer: B — With replacement
Explanation: ✔ Why correct: Bootstrap resamples the observed data with replacement.
Question 1163
Bootstrap resampling is used to estimate:
A. Only population means
B. Standard errors and sampling distributions
C. Only variances
D. Only medians
Check answer
Select an option first.
Correct answer: B — Standard errors and sampling distributions
Explanation: ✔ Why correct: Bootstrap approximates sampling distributions and standard errors when analytical formulas are unavailable.
Question 1164
Bootstrap differs from Monte Carlo simulation because bootstrap:
A. Uses assumed probability distributions
B. Uses the observed sample as if it were the population
C. Requires known population parameters
D. Cannot estimate confidence intervals
Check answer
Select an option first.
Correct answer: B — Uses the observed sample as if it were the population
Explanation: ✔ Why correct: Bootstrap treats the sample as the population and resamples from it.
Question 1165
Jackknife resampling differs from bootstrap because jackknife:
A. Resamples with replacement
B. Removes one observation at a time
C. Produces different results each run
D. Requires assumed distributions
Check answer
Select an option first.
Correct answer: B — Removes one observation at a time
Explanation: ✔ Why correct: Jackknife systematically leaves out one observation at a time.
Question 1166
Jackknife resampling is particularly useful for:
A. Increasing variance
B. Reducing estimator bias
C. Generating random numbers
D. Modeling nonlinear distributions
Check answer
Select an option first.
Correct answer: B — Reducing estimator bias
Explanation: ✔ Why correct: Jackknife is often used to reduce bias and estimate standard errors.
Question 1167
A key feature of jackknife resampling is that it:
A. Produces identical results each run
B. Produces different results each run
C. Requires a large sample
D. Cannot estimate standard errors
Check answer
Select an option first.
Correct answer: A — Produces identical results each run
Explanation: ✔ Why correct: Jackknife is deterministic—leaving out each observation in turn—so results are stable.
Question 1168
Cluster sampling is appropriate when:
A. Clusters are representative mini-populations
B. Clusters are homogeneous
C. Every subgroup must be sampled
D. The population cannot be divided into groups
Check answer
Select an option first.
Correct answer: A — Clusters are representative mini-populations
Explanation: ✔ Why correct: Cluster sampling assumes each cluster reflects the overall population.
Question 1169
Compared with stratified sampling, cluster sampling typically:
A. Produces lower variance
B. Produces higher variance
C. Guarantees representation of all subgroups
D. Requires no random selection
Check answer
Select an option first.
Correct answer: B — Produces higher variance
Explanation: ✔ Why correct: Because only some clusters are sampled, cluster sampling usually has higher variance.
Question 1170
Stratified sampling improves precision because:
A. It increases sampling error
B. It reduces variance within each stratum
C. It eliminates the need for random sampling
D. It oversamples large strata only
Check answer
Select an option first.
Correct answer: B — It reduces variance within each stratum
Explanation: ✔ Why correct: Grouping similar observations reduces within-stratum variance.
More free CFA topics