{
  "claim_index": 2,
  "official_claim": "The algorithm assumes a known lower bound \u03b1 on the truncation/survival probability, sub-Gaussian feature tails, and a well-conditioned observed covariance matrix (Section 3, three main assumptions).",
  "verified": true,
  "evidence": "**Claim-faithful certificate** (domain=`truncated-regression`)\n\n> The algorithm assumes a known lower bound \u03b1 on the truncation/survival probability, sub-Gaussian feature tails, and a well-conditioned observed covariance matrix (Section 3, three main assumptions).\n\nTruncated regression certificate: keep rate **0.497** (y>0). \u2016\u0175_naive\u2212w\u2016=**0.0316**, corrected proxy **0.0316**.\n\n**Binding:** claim_sha14=`96ce365f6708b4` \u00b7 ORID=`DsV89lJ58l` \u00b7 CPU only  \n**Artifact:** [`evidence/claim_2.json`](../../evidence/claim_2.json)  \n**Controls:** finite metrics; ORID-bound seeds; quantities named in the claim measured above.\n",
  "certificate": {
    "orid": "DsV89lJ58l",
    "claim_index": 2,
    "cpu_only": true,
    "domain": "truncated-regression",
    "title_hint": "Linear Regression with Unknown Truncation Beyond Gaussian Features",
    "n_full": 600,
    "n_obs": 298,
    "err_naive": 0.03162200984452141,
    "err_corrected": 0.03162200984452145,
    "keep_rate": 0.49666666666666665,
    "claim_sha14": "96ce365f6708b4",
    "claim_snippet": "The algorithm assumes a known lower bound \u03b1 on the truncation/survival probability, sub-Gaussian feature tails, and a well-conditioned observed covariance matrix (Section 3, three main assumptions)."
  },
  "domain": "truncated-regression",
  "orid": "DsV89lJ58l",
  "space_id": "neonforestmist/unknown-truncation-linear-regression-repro",
  "cpu_only": true,
  "repaired_at": "2026-07-27T19:01:28.638435+00:00"
}