{
  "id": 1696,
  "title": "Question regarding rule # 9",
  "url": "/competitions/benchmark-bond-trade-price-challenge/discussion/1696",
  "author_name": "",
  "post_date": "2012-04-08T06:23:32.723Z",
  "votes": null,
  "comment_count": 1,
  "views": 2246,
  "content": "<p>Rule # 9 states:</p>\r\n<p>Participants will not use data other than that provided to estimate their model.</p>\r\n<p>Does this apply to the&nbsp;original training and testing sets?&nbsp; If I read Dan's post correctly, the &quot;new&quot; training set is from a different time period than the &quot;old&quot; training set.&nbsp; Does the model have to be trained exclusively from the &quot;new&quot; training/testing\r\n sets or can it use both the old and the new?</p>",
  "messages": [
    {
      "id": "10015",
      "postDate": "04/08/2012 06:23:32",
      "content": "<p>Rule # 9 states:</p>\r\n<p>Participants will not use data other than that provided to estimate their model.</p>\r\n<p>Does this apply to the&nbsp;original training and testing sets?&nbsp; If I read Dan's post correctly, the &quot;new&quot; training set is from a different time period than the &quot;old&quot; training set.&nbsp; Does the model have to be trained exclusively from the &quot;new&quot; training/testing\r\n sets or can it use both the old and the new?</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "10067",
      "postDate": "04/11/2012 13:56:30",
      "content": "<p>I would advise against this. &nbsp;While the data is superficially similar, there are small differences that might affect your results. &nbsp;The universe of bonds selected is somewhat different. &nbsp;The curve based price is improved in subtle ways. &nbsp;Trades are ignored\r\n as outliers/errors in different ways. &nbsp;There might be other differences as well.</p>\r\n<p>-Dan</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 10067,
      "author_name": "danglaser",
      "author_url": "",
      "post_date": "04/11/2012 13:56:30",
      "content": "<p>I would advise against this. &nbsp;While the data is superficially similar, there are small differences that might affect your results. &nbsp;The universe of bonds selected is somewhat different. &nbsp;The curve based price is improved in subtle ways. &nbsp;Trades are ignored\r\n as outliers/errors in different ways. &nbsp;There might be other differences as well.</p>\r\n<p>-Dan</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "10015": "",
    "10067": ""
  },
  "source": "meta"
}