{
  "aliases": null,
  "backlinks": [
    {
      "field": "defines",
      "target": "JEAYPXPD87"
    }
  ],
  "body": "\n## Summary\n\nThe Akaike information criterion (AIC) scores a fitted statistical model by its maximized [[likelihood|0VW9E4QXR6]], with a penalty for the number of parameters.\nAmong candidate models, one prefers the model with the smallest AIC.\nIt rests on an estimate of the expected [[Kullback–Leibler divergence|PQBW43X5RE]] between the true law and the fitted model, and it aims at good prediction.\n\n## Explanations\n\n### L1 human\n\nA model with more parameters always fits the data at hand at least as well, but it may only be fitting noise.\nAIC balances the fit, measured by the likelihood, against the complexity, measured by the number of parameters.\nEach extra parameter costs $2$ units, so it must raise twice the log-likelihood by more than $2$ to be worth adding.\nFor example, a cubic regression is preferred to a straight line only if its better fit pays for its two extra coefficients.\n",
  "commit": "ff8a40749dc16545e3d2fc637f6b0196e81d0bfa",
  "front": {
    "kind": "concept",
    "msc": [
      "62B10"
    ],
    "name": "Akaike information criterion",
    "same_as": {
      "wikidata": "Q1662573",
      "wikipedia": "Akaike information criterion"
    },
    "sources": [
      {
        "url": "https://en.wikipedia.org/wiki/Akaike_information_criterion"
      }
    ],
    "tag": "CMAX3KF9H0"
  },
  "kind": "concept",
  "link": "https://tamnd.github.io/math2/node/CMAX3KF9H0/",
  "links": null,
  "markdown": "https://tamnd.github.io/math2/node/CMAX3KF9H0.md",
  "msc": [
    "62B10"
  ],
  "name": "Akaike information criterion",
  "prov": null,
  "role": "",
  "source": "https://github.com/tamnd/math2/blob/main/nodes/CMA/CMAX3KF9H0.md",
  "status": null,
  "tag": "CMAX3KF9H0",
  "uri": "m2://CMAX3KF9H0",
  "url": "https://tamnd.github.io/math2/wiki/akaike-information-criterion/",
  "writing": {
    "tag": "CMAX3KF9H0",
    "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
      }
    ]
  }
}