{
  "aliases": null,
  "backlinks": [
    {
      "field": "defines",
      "target": "QXDY7STQWG"
    }
  ],
  "body": "\n## Summary\n\nA missing data mechanism is the probability law that decides which entries of a data set are not observed.\nThe standard classification has three classes: missing completely at random (MCAR), missing at random (MAR) and missing not at random (MNAR).\nThe class decides which analyses stay valid.\nUnder MAR, if the mechanism has its own separate parameters, likelihood and Bayesian inference can ignore it.\nUnder MNAR the mechanism must be modelled.\n\n## Explanations\n\n### L1 human\n\nData are often incomplete: people skip survey questions, sensors fail, patients leave a study.\nThe key question is whether the fact that a value is missing is linked to the values themselves.\nIf a lab machine breaks at random times, the missing values are MCAR, and the rest of the data is still a fair sample.\nIf older people skip an income question more often and age is recorded, the data are MAR.\nIf people with high incomes skip the income question because their income is high, the data are MNAR, and a naive average of the answers is too low.\n",
  "commit": "ff8a40749dc16545e3d2fc637f6b0196e81d0bfa",
  "front": {
    "kind": "concept",
    "msc": [
      "62D10"
    ],
    "name": "Missing data mechanisms",
    "sources": [
      {
        "url": "https://en.wikipedia.org/wiki/Missing_data"
      }
    ],
    "tag": "ES4MC30ZES"
  },
  "kind": "concept",
  "link": "https://tamnd.github.io/math2/node/ES4MC30ZES/",
  "links": null,
  "markdown": "https://tamnd.github.io/math2/node/ES4MC30ZES.md",
  "msc": [
    "62D10"
  ],
  "name": "Missing data mechanisms",
  "prov": null,
  "role": "",
  "source": "https://github.com/tamnd/math2/blob/main/nodes/ES4/ES4MC30ZES.md",
  "status": null,
  "tag": "ES4MC30ZES",
  "uri": "m2://ES4MC30ZES",
  "url": "https://tamnd.github.io/math2/wiki/missing-data-mechanisms/",
  "writing": {
    "tag": "ES4MC30ZES",
    "kind": "concept",
    "level": "W0",
    "next": [
      {
        "what": "an inventory row",
        "note": "the row of the inventory that the node accounts for",
        "have": 0,
        "want": 1
      },
      {
        "what": "a source with a locator",
        "have": 0,
        "want": 1
      }
    ]
  }
}