Test your knowledge of bayesian vs.
In Bayesian analysis, a 95 % interval for the lift is called a ______ interval.
prediction
credible
confidence
tolerance
A credible interval contains 95 % of the posterior probability mass for the parameter, reflecting degree of belief.
Frequentist p‑values describe the probability of obtaining data as extreme assuming the ______ is true.
null hypothesis
posterior mean
prior
alternative hypothesis
They do not give the probability that the null itself is true but the likelihood of the observed statistic under it.
Bayesian methods naturally incorporate previous evidence through the use of a ______ distribution.
prior
sampling
F‑test
likelihood
Priors encode earlier knowledge or assumptions and update with new data to form the posterior.
Stopping rules (peeking) affect Frequentist validity but, under a proper likelihood, Bayesian posteriors are ______ to when you stop.
undefined
highly sensitive
invariant
unreliable
Bayesian updating applies Bayes’ theorem at any point, so the posterior remains coherent irrespective of sampling plan.
The Bayes Factor compares the probability of the data under two competing ______.
models / hypotheses
segments
confidence intervals
priors
It is a ratio of marginal likelihoods and serves as evidence in favour of one hypothesis over another.
A Frequentist 95 % confidence interval will capture the true parameter 95 % of the time ______.
over repeated identical experiments
after applying priors
for the single observed dataset
only with large samples
The interpretation is long‑run and not a probability statement about the parameter in a specific test.
Using non‑informative priors in Bayesian tests tends to make results more similar to ______ outcomes.
Frequentist
sequential
clustered
bootstrapped
A flat prior contributes little information, so the posterior is dominated by data, converging on likelihood‑based estimates akin to classical methods.
The probability that variant B is better than control A calculated from the posterior is often referred to as ______.
power
probability to beat control (PBTC)
effect size d
alpha value
PBTC offers a direct, intuitive metric for decision‑making, unlike p‑values which do not give directional probabilities.
Conjugate prior‑likelihood pairs are prized in Bayesian testing because they yield posterior distributions of the ______ functional form.
Gaussian
unknown
same
opposite
Mathematical conjugacy simplifies computation, enabling closed‑form updates that are fast for high‑traffic experimentation.
Bayesian decision theory can incorporate utility functions to optimise for business value, whereas classical tests typically stop at ______.
credible mass
statistical significance
expected loss
posterior risk
Bayesian frameworks connect probability with payoff, allowing decisions that maximise expected utility rather than p‑values alone.
Starter
Brush up on the basics and avoid costly mistakes.
Solid
You know your stuff—refine a few nuances for mastery.
Expert!
Outstanding! You could teach this topic.
