{
  "aliases": null,
  "backlinks": [
    {
      "field": "defines",
      "target": "AMXXM288Y0"
    }
  ],
  "body": "\n## Summary\n\nA minimal sufficient statistic is a [[sufficient statistic|WP4TGXJQJG]] that reduces the data as much as possible without losing information about the parameter.\nIt is a function of every other sufficient statistic.\nIn regular models it exists and can be found from ratios of likelihoods.\n\n## Explanations\n\n### L1 human\n\nThe whole data set is always sufficient, but it is not a summary at all.\nA minimal sufficient statistic is the most compressed summary that still keeps everything relevant about the parameter.\nFor coin tosses it is the number of heads; for normal data with unknown mean and variance it is the pair (sample mean, sample variance).\nAny further compression would lose information.\n",
  "commit": "ff8a40749dc16545e3d2fc637f6b0196e81d0bfa",
  "front": {
    "kind": "concept",
    "msc": [
      "62B05"
    ],
    "name": "Minimal sufficient statistic",
    "sources": [
      {
        "locator": "Minimal sufficiency",
        "url": "https://en.wikipedia.org/wiki/Sufficient_statistic"
      }
    ],
    "specializes": [
      "WP4TGXJQJG"
    ],
    "tag": "RADE160VJN"
  },
  "kind": "concept",
  "link": "https://tamnd.github.io/math2/node/RADE160VJN/",
  "links": [
    {
      "field": "specializes",
      "target": "WP4TGXJQJG"
    }
  ],
  "markdown": "https://tamnd.github.io/math2/node/RADE160VJN.md",
  "msc": [
    "62B05"
  ],
  "name": "Minimal sufficient statistic",
  "prov": null,
  "role": "",
  "source": "https://github.com/tamnd/math2/blob/main/nodes/RAD/RADE160VJN.md",
  "status": null,
  "tag": "RADE160VJN",
  "uri": "m2://RADE160VJN",
  "url": "https://tamnd.github.io/math2/wiki/minimal-sufficient-statistic/",
  "writing": {
    "tag": "RADE160VJN",
    "kind": "concept",
    "level": "W0",
    "next": [
      {
        "what": "an inventory row",
        "note": "the row of the inventory that the node accounts for",
        "have": 0,
        "want": 1
      }
    ]
  }
}