Zero-inflated Poisson distribution Created 2026-10-03 Updated 2026-11-05
A zero-inflated Poisson distribution is a mixture of a point mass at zero and a Poisson distribution. If the Poisson component has weight and mean , then its mean is and its variance is .
Zero-point motion Created 2026-10-18 Updated 2026-11-05
Zero-point motion is the nonzero quantum fluctuation of position and momentum in a system's ground state. For a quantum harmonic oscillator, the ground state has energy and nonzero position and momentum variances even at zero temperature. It does not represent a thermal mixture and need not carry entropy. The large zero-point motion of light helium atoms helps explain why helium can remain a liquid at low pressure even as temperature approaches absolute zero.
When a full-rank normal regression with positive residual degrees of freedom happens to fit all supplied observations exactly, its residual estimate of variance is zero. The ordinary Studentized prediction-interval formula formally collapses, but the statistic divides by zero. Under positive true Gaussian variance such samples form a null set; exact rounded or constructed data must not be treated as proof that future randomness vanishes.
Zero-truncated binomial distribution Created 2026-10-14 Updated 2026-11-05
The zero-truncated binomial distribution has probabilities for . Its natural parameter is and its cumulant is . mean and variance are and , where . The conditional normalizer depends on , so its likelihood is not proportional to the untruncated likelihood even after a positive observation. At and , its likelihood supremum is at rather than at .
If a random variable has variance zero and mean , it equals almost surely. For a target probability strictly between zero and one, meets the target exactly when . Substituting zero variance into a positive-variance non-strict quantile constraint incorrectly includes , where the overflow probability is one. The distinction comes from the atom at the threshold.