{
  "slug": "confidence-sequence",
  "term": "Confidence sequence",
  "aliases": [
    "time-uniform confidence sequence",
    "anytime-valid confidence sequence"
  ],
  "definition": "A confidence sequence is a sequence of confidence sets whose simultaneous coverage of the target at all monitored times is at least a stated level, under specified assumptions.",
  "note": "For a fixed target theta, the guarantee is P(theta belongs to C_t for every t >= 1) >= 1 - alpha. Separate coverage at each fixed sample size is weaker. Optional stopping can use observations collected so far without invalidating a suitable confidence sequence. [ramdas-savi]\n\nTime-uniform coverage also protects selecting an earlier interval from the same valid sequence after seeing the history: miscoverage at that selected time implies miscoverage at some time. This is unconditional coverage over repeated procedures, not a posterior probability or guaranteed coverage conditional on a favorable stopping signal. [howard-cs]\n\nSelecting a favorable method or device-model target after inspecting results needs separate justification. The observation model still matters: a method calibrated for independent, constant-risk endpoints does not automatically handle shared failures, censoring or changing populations. [ramdas-savi]\n\nA finite set of planned looks can instead allocate error levels whose sum is at most alpha. This controls that set by the union bound without independent looks; it does not by itself authorize unlimited further looks. [nist-bonferroni]",
  "related": [
    "confidence-interval",
    "binomial-proportion",
    "p-value"
  ],
  "citations": [
    {
      "id": "howard-cs",
      "title": "Howard et al. (2021): Time-uniform, nonparametric, nonasymptotic confidence sequences; equation 1, Lemma 3",
      "url": "https://arxiv.org/html/1810.08240v6",
      "accessed": "2026-11-19"
    },
    {
      "id": "ramdas-savi",
      "title": "Ramdas et al. (2023): Game-Theoretic Statistics and Safe Anytime-Valid Inference; sections 2.8 and 6.4",
      "url": "https://arxiv.org/html/2210.01948v2",
      "accessed": "2026-11-19"
    },
    {
      "id": "nist-bonferroni",
      "title": "NIST e-Handbook 7.4.7.3: Bonferroni's method, general inequality",
      "url": "https://www.itl.nist.gov/div898/handbook/prc/section4/prc473.htm",
      "accessed": "2026-11-19"
    }
  ],
  "status": "research",
  "advantages": [],
  "limitations": [
    "Time-uniform validity does not require the interval width to shrink to zero. [howard-cs]"
  ],
  "provenance": [
    "Methodological definition, not a device measurement or an implemented confidence-sequence estimator. The related AFR analysis retains its separate exact finite-look example."
  ],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-11-19"
}
