{
  "id": 1339,
  "title": "Rules Change with Relaunch",
  "url": "/competitions/benchmark-bond-trade-price-challenge/discussion/1339",
  "author_name": "",
  "post_date": "2012-02-07T16:36:45.657Z",
  "votes": 1,
  "comment_count": 1,
  "views": 1790,
  "content": "<p>With the relaunch of this competition, we have made the rules more permissive. &nbsp;Specifically, the training data is now in order and contains bond_id to aid you in reconstructing a full time series per bond if you so desire. &nbsp;The training data also includes\r\n every trade as a row, so each row contains the previous row as its most recent lagged trade.</p>\r\n<p>The test data contains no bond_id, is not in order, and does not have overlapping rows.</p>\r\n<p>We believe that this change makes the rules simpler, easier to enforce and more logical. &nbsp;Please let me know if you have any questions or concerns.</p>\r\n<p>-Dan</p>",
  "messages": [
    {
      "id": "8470",
      "postDate": "02/07/2012 16:36:45",
      "content": "<p>With the relaunch of this competition, we have made the rules more permissive. &nbsp;Specifically, the training data is now in order and contains bond_id to aid you in reconstructing a full time series per bond if you so desire. &nbsp;The training data also includes\r\n every trade as a row, so each row contains the previous row as its most recent lagged trade.</p>\r\n<p>The test data contains no bond_id, is not in order, and does not have overlapping rows.</p>\r\n<p>We believe that this change makes the rules simpler, easier to enforce and more logical. &nbsp;Please let me know if you have any questions or concerns.</p>\r\n<p>-Dan</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "8491",
      "postDate": "02/08/2012 21:27:10",
      "content": "<p>The evaluation tab explains this rather well, but the upshot is that we use mean absolute error weighted by the square root of the time since the last trade. &nbsp;This is to avoid the most often traded bonds (which are also in many cases the &quot;easiest&quot;) dominating\r\n the importance.</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 8491,
      "author_name": "danglaser",
      "author_url": "",
      "post_date": "02/08/2012 21:27:10",
      "content": "<p>The evaluation tab explains this rather well, but the upshot is that we use mean absolute error weighted by the square root of the time since the last trade. &nbsp;This is to avoid the most often traded bonds (which are also in many cases the &quot;easiest&quot;) dominating\r\n the importance.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "8470": "",
    "8491": ""
  },
  "source": "meta"
}