{
  "id": 327602,
  "title": "Final/Private dataset evaluation on October 2019 statement",
  "url": "/competitions/amex-default-prediction/discussion/327602",
  "author_name": "Seeker",
  "post_date": "2022-05-28T03:53:44.438000",
  "votes": 42,
  "comment_count": 5,
  "views": 0,
  "content": "<p>For test dataset, distribution of performance vintage is as below,</p>\n<p><a href=\"https://www.kaggle.com/code/taran8727/exploration-time-periods-and-credit-defaults\" target=\"_blank\">Details here</a> </p>\n<p>Last Statement Month</p>\n<p>2019-04-01     467966  - 50.6%<br>\n2019-10-01     456655   - 49.4%</p>\n<p>As per information on leader board  </p>\n<blockquote>\n  <p>This leaderboard is calculated with approximately 51% of the test data. The final results will be based on the other 49%, so the final standings may be different.</p>\n</blockquote>\n<p>From this it seems, <br>\n<strong>Private set (49%) is from October 2019 vintage</strong> (Now confirmed by <a href=\"https://www.kaggle.com/celiker\" target=\"_blank\">@celiker</a> ) </p>",
  "messages": [
    {
      "id": 1803654,
      "postDate": "2022-05-28T03:53:44.440Z",
      "content": "<p>For test dataset, distribution of performance vintage is as below,</p>\n<p><a href=\"https://www.kaggle.com/code/taran8727/exploration-time-periods-and-credit-defaults\" target=\"_blank\">Details here</a> </p>\n<p>Last Statement Month</p>\n<p>2019-04-01     467966  - 50.6%<br>\n2019-10-01     456655   - 49.4%</p>\n<p>As per information on leader board  </p>\n<blockquote>\n  <p>This leaderboard is calculated with approximately 51% of the test data. The final results will be based on the other 49%, so the final standings may be different.</p>\n</blockquote>\n<p>From this it seems, <br>\n<strong>Private set (49%) is from October 2019 vintage</strong> (Now confirmed by <a href=\"https://www.kaggle.com/celiker\" target=\"_blank\">@celiker</a> ) </p>",
      "rawMarkdown": "For test dataset, distribution of performance vintage is as below,\n\n[Details here](https://www.kaggle.com/code/taran8727/exploration-time-periods-and-credit-defaults) \n\nLast Statement Month\n\n2019-04-01 \t467966 \t- 50.6%\n2019-10-01 \t456655   - 49.4%\n\nAs per information on leader board  \n> This leaderboard is calculated with approximately 51% of the test data. The final results will be based on the other 49%, so the final standings may be different.\n\nFrom this it seems, \n**Private set (49%) is from October 2019 vintage** (Now confirmed by @celiker ) \n\n",
      "votes": 40
    },
    {
      "id": 1804670,
      "postDate": "2022-05-29T10:41:23.190Z",
      "content": "<p>I just tested(changed the predictions of that 49%, for an existing submission) and the score didn't change. So this is probably true. Private lb is from the ones with the last statement in 2019 October</p>",
      "rawMarkdown": "I just tested(changed the predictions of that 49%, for an existing submission) and the score didn't change. So this is probably true. Private lb is from the ones with the last statement in 2019 October",
      "votes": 8,
      "replies": [
        {
          "id": 1805163,
          "postDate": "2022-05-29T21:01:53.187Z",
          "rawMarkdown": "",
          "isDeleted": true
        }
      ]
    },
    {
      "id": 1840669,
      "postDate": "2022-07-02T13:28:10.267Z",
      "content": "<p>So basically the target is calculated based on:</p>\n<ul>\n<li>For Public: deliquesceny event in the 18 months following 2019-04</li>\n<li>For Private: deliquesceny event in the 18 months following 2019-10</li>\n</ul>\n<p>Correct?</p>",
      "rawMarkdown": "So basically the target is calculated based on:\n- For Public: deliquesceny event in the 18 months following 2019-04\n- For Private: deliquesceny event in the 18 months following 2019-10\n\nCorrect?",
      "votes": 1
    },
    {
      "id": 1805287,
      "postDate": "2022-05-30T03:35:58.183Z",
      "content": "<p>Thanks Seeker. This is a helpful and great discovery!</p>",
      "rawMarkdown": "Thanks Seeker. This is a helpful and great discovery!",
      "votes": 2
    },
    {
      "id": 1805082,
      "postDate": "2022-05-29T19:02:49.247Z",
      "content": "<p>Thank you for confirming, </p>",
      "rawMarkdown": "Thank you for confirming, \n"
    }
  ],
  "comments": [
    {
      "id": 1804670,
      "author_name": "Cihat Emre Çeliker",
      "author_url": "",
      "post_date": "2022-05-29T10:41:23.190000",
      "content": "<p>I just tested(changed the predictions of that 49%, for an existing submission) and the score didn't change. So this is probably true. Private lb is from the ones with the last statement in 2019 October</p>",
      "votes": 8,
      "replies": [
        {
          "id": 1805163,
          "author_name": "",
          "author_url": "",
          "post_date": "2022-05-29T21:01:53.187000",
          "content": "",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 1840669,
      "author_name": "Gad Benram",
      "author_url": "",
      "post_date": "2022-07-02T13:28:10.267000",
      "content": "<p>So basically the target is calculated based on:</p>\n<ul>\n<li>For Public: deliquesceny event in the 18 months following 2019-04</li>\n<li>For Private: deliquesceny event in the 18 months following 2019-10</li>\n</ul>\n<p>Correct?</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 1805287,
      "author_name": "Chris Deotte",
      "author_url": "",
      "post_date": "2022-05-30T03:35:58.183000",
      "content": "<p>Thanks Seeker. This is a helpful and great discovery!</p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 1805082,
      "author_name": "Seeker",
      "author_url": "",
      "post_date": "2022-05-29T19:02:49.247000",
      "content": "<p>Thank you for confirming, </p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1803654": "For test dataset, distribution of performance vintage is as below,\n\n[Details here](https://www.kaggle.com/code/taran8727/exploration-time-periods-and-credit-defaults) \n\nLast Statement Month\n\n2019-04-01 \t467966 \t- 50.6%\n2019-10-01 \t456655   - 49.4%\n\nAs per information on leader board  \n> This leaderboard is calculated with approximately 51% of the test data. The final results will be based on the other 49%, so the final standings may be different.\n\nFrom this it seems, \n**Private set (49%) is from October 2019 vintage** (Now confirmed by @celiker ) \n\n",
    "1804670": "I just tested(changed the predictions of that 49%, for an existing submission) and the score didn't change. So this is probably true. Private lb is from the ones with the last statement in 2019 October",
    "1840669": "So basically the target is calculated based on:\n- For Public: deliquesceny event in the 18 months following 2019-04\n- For Private: deliquesceny event in the 18 months following 2019-10\n\nCorrect?",
    "1805287": "Thanks Seeker. This is a helpful and great discovery!",
    "1805082": "Thank you for confirming, \n"
  }
}