Let be fixed and known, and let be independent and identically distributed Bernoulli random variables with common success probability . Their joint probability mass function, on the binary sample space , isThere are arrangements with successes, so has the binomial distribution. For and , its exact conditional full-data probability mass function isThis conditional distribution is uniform over arrangements, with no dependence on : the sum is a sufficient statistic. [ref]
Sufficiency (Stat 210a, Fall 2025) by Will Fithian. Sufficiency, Will Fithian, “Independent Bernoulli sequence”: conditional pmf
Four Bernoulli observations grouped by their sufficient sum
. All sixteen binary sequences are grouped by . Within each group the conditional distribution is uniform; changing changes the group's total probability, not the conditional weights within it. Details.For , the group sizes are and the within-group conditional probabilities are respectively . For example, leaves exactly , each with conditional probability . At , each such sequence has unconditional probability and the group has probability . At , these values are and ; at they return to and . The conditional weights stay in all three cases.
At the boundary , only has positive probability; at , only does. Dividing by on the other groups would divide by zero. Nevertheless, the same uniform-on-arrangements Markov kernel is a valid common version for the entire family : it gives the correct point mass on the one observed group at each boundary and assigns permissible versions on its unobserved groups. Define the joint probability mass function there by the original product of the one-trial Bernoulli probability mass functions, or use the limiting convention in the compact formula.
The sample mean is an unbiased estimator of and retains the sufficient statistic because is known. By contrast, is also an unbiased estimator of but, for , is not sufficient: the conditional success probability of the independent given is still . Thus unbiasedness does not imply sufficiency.
Sufficiency also does not imply completeness of a statistic. For , the full-data sufficient statistic admits the nonzero function with expectation zero for every . Nor does minimality imply completeness of a statistic: for and the restricted statistical model , is a minimal sufficient statistic since the likelihood ratio distinguishes its three values. Yet has expectation at both parameter values and is never zero. Changing the parameter family can change completeness of a statistic even while preserving sufficiency.
For independent binary edges with log odds , the natural statistics are the sum of the vertex degrees in each class and the total number of within-class edges. They form a minimal sufficient statistic whenever the natural parameter space contains an open subset of .
Past exam of the mathematics course of the University of Cambridge 2010 ii Paper 4 27J Solution Created 2026-09-24 Updated 2026-10-14
A complete statistic for a family if every measurable with and for all satisfies almost surely under every . It is boundedly complete if the same implication is required only for bounded measurable . Completeness therefore implies bounded completeness.
The vector is multivariate normal with mean zero and covarianceIts determinant is and its inverse is . Hence the joint density isThe Fisher-Neyman factorization theorem gives sufficiency. For minimality use the likelihood-ratio criterion for minimal sufficiency. The density ratio for is independent of precisely whenis constant. Multiplying by shows that its constant would produce a quadratic term on the right whereas the left is linear, so that constant is zero. The remaining coefficients give . Thus is minimal sufficient statistic.
Each is standard normal, so . The integrable, nonzero function has zero expectation for every parameter, proving that is not complete.
By sufficiency, choose a version common to all parameters. The tower property givesOn the event we necessarily have , so there almost surely. This event has positive probability under every parameter because the joint normal density is strictly positive on a ball about the origin. Thus is not identically . The bounded function has zero expectation for every parameter but is not almost surely zero. Consequently
Past exam of the mathematics course of the University of Cambridge 2016 ib Paper 1 19H c Solution Created 2026-09-24 Updated 2026-10-06
Write , and . The joint probability density factors asBy the Fisher-Neyman factorization theorem, is a sufficient statistic. The same factorization shows that the smaller statistic is also a sufficient statistic. This gives the sufficient statistic for a symmetric uniform sample.
In fact, the likelihood-ratio criterion for minimal sufficiency identifies as a minimal sufficient statistic: two sample points with the same have identical likelihoods for all ; with different , their likelihoods have different parameter supports, and so are not proportional as functions of . Equivalently, the condition that for every and some positive constant holds exactly when their absolute maxima agree.
To see rigorously that cannot be recovered from almost surely, fix any . The sample has a distribution invariant under simultaneous sign reversal. If almost surely, the same identity would hold for the sign-reversed sample on another probability-one set. Since is unchanged by sign reversal, it would follow thatand hence almost surely. But this event has probability zero for a finite sample from a continuous uniform distribution: it entails for some pair of indices, including the possible case . Therefore is not a function of the sufficient statistic . The pair of extrema is not minimally sufficient, for every , even though it is sufficient.
Past exam of the mathematics course of the University of Cambridge 2017 ib Paper 1 19H a Solution Created 2026-09-24 Updated 2026-10-05
A sufficient statistic has a conditional distribution of the full data given its value that is independent of the unknown parameter. A minimal sufficient statistic is sufficient and is a function of every other sufficient statistic, up to null sets. For a dominated family with common positive support, the likelihood-ratio criterion for minimal sufficiency states that is minimal sufficient if holds exactly when is independent of .
For , the joint probability mass function isThe Fisher-Neyman factorization theorem proves sufficiency of . For two samples, the likelihood ratio is , independent of exactly when their sums agree. Thus is minimal sufficient. Any one-to-one transformation of it is also minimal sufficient, so chooseThis chosen estimator is an unbiased estimator, since each observation has mean . If instead the statistic is reported as , its expectation is , and as an estimator of it has bias (zero only when ). Being a minimal sufficient statistic is unchanged by this rescaling; being an unbiased estimator is not.
Past exam of the mathematics course of the University of Cambridge 2020 ib Paper 1 19H b Solution Created 2026-09-24 Updated 2026-09-29
A statistic is sufficient for if the conditional distribution of the full sample given does not depend on . It is minimal sufficient if it is a function of every sufficient statistic.
The likelihood factors asso the Fisher-Neyman factorization theorem shows that , and hence its one-to-one transform , is sufficient. Moreover, for two samples and , the ratio is independent of exactly when their maxima agree. The likelihood-ratio criterion for minimal sufficiency therefore shows that and are minimal sufficient.
For , the sample mean is not sufficient: two samples can have the same mean but different maxima, and their likelihood ratio then depends on through the support indicators. It is consequently not minimal sufficient. For the degenerate special case , the sample mean and maximum coincide and both conclusions reverse.
Sufficient statistic Created 2026-09-24 Updated 2026-10-14
A statistic is sufficient for a statistical model when the conditional distribution of the full data given has a version independent of the unknown statistical parameter . Once is recorded, the remaining uncertainty about the data's arrangement can be generated without knowing . [ref]
Sufficiency (Stat 210a, Fall 2025) by Will Fithian. Sufficiency, Will Fithian, “Sufficiency”: definition
For data and statistic taking values in standard Borel spaces, express this precisely with one Markov kernel , common to every statistical parameter value:for each measurable event . Thus is a regular conditional distribution of given . Equality is only required outside a null set for the distribution of under each ; the choice of version matters on impossible conditioning events. For a continuous probability distribution, the event may have zero probability even at meaningful observations, so the definition uses a regular conditional distribution, rather than division by .
A sufficient statistic is a property of a specified statistical model, rather than of a formula for summarizing data in isolation. It need not be an unbiased estimator or even shorten the record: the full data is always sufficient. An estimator instead targets a quantity such as a mean. An ancillary statistic has a parameter-independent marginal probability distribution, whereas a sufficient statistic has a parameter-independent conditional full-data law. A minimal sufficient statistic permits the greatest reduction among sufficient statistics. A complete statistic satisfies a separate condition about zero expectations; a complete sufficient statistic must satisfy both conditions.
Sufficient statistic for a symmetric uniform sample Created 2026-10-06 Updated 2026-11-05
For independent samples uniform on with , the absolute maximum is a minimal sufficient statistic, since the likelihood is . The pair of sample minimum and maximum is a sufficient statistic but is not minimal: simultaneous sign reversal preserves and generally changes that pair.
