{
  "id": 285713,
  "title": "How was the data created? ",
  "url": "/competitions/g-research-crypto-forecasting/discussion/285713",
  "author_name": "",
  "post_date": "2021-11-05T22:15:48.917638500Z",
  "votes": 4,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees. I wonder if the competition host can share some insight about the dataset creation process?</p>\n<p>I am trying to build a system that can automatically pull in trade data after 2021-09-21 which is then properly version controlled and easily distributed with fellow Kagglers. Thought of doing this since public data is allowed to use. </p>\n<p>One major bottleneck is the data creation process. Any help will be appreciated. If you all think this is not the right path to follow, will love to know your thoughts. </p>\n<h3>Note</h3>\n<p>Since the competition's LB score is of no use, if the data creation process is made clear, we can build our own data and evaluate the model on it. I think it will help come up with a more useful solution. </p>\n<p>Tagging <a href=\"https://www.kaggle.com/andrewscherer\" target=\"_blank\">@andrewscherer</a> for visibility. </p>",
  "messages": [
    {
      "id": "1572722",
      "postDate": "11/05/2021 22:15:48",
      "content": "<p>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees. I wonder if the competition host can share some insight about the dataset creation process?</p>\n<p>I am trying to build a system that can automatically pull in trade data after 2021-09-21 which is then properly version controlled and easily distributed with fellow Kagglers. Thought of doing this since public data is allowed to use. </p>\n<p>One major bottleneck is the data creation process. Any help will be appreciated. If you all think this is not the right path to follow, will love to know your thoughts. </p>\n<h3>Note</h3>\n<p>Since the competition's LB score is of no use, if the data creation process is made clear, we can build our own data and evaluate the model on it. I think it will help come up with a more useful solution. </p>\n<p>Tagging <a href=\"https://www.kaggle.com/andrewscherer\" target=\"_blank\">@andrewscherer</a> for visibility. </p>",
      "rawMarkdown": "I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees. I wonder if the competition host can share some insight about the dataset creation process?\n\nI am trying to build a system that can automatically pull in trade data after 2021-09-21 which is then properly version controlled and easily distributed with fellow Kagglers. Thought of doing this since public data is allowed to use. \n\nOne major bottleneck is the data creation process. Any help will be appreciated. If you all think this is not the right path to follow, will love to know your thoughts. \n\n### Note\n\nSince the competition's LB score is of no use, if the data creation process is made clear, we can build our own data and evaluate the model on it. I think it will help come up with a more useful solution. \n\nTagging @andrewscherer for visibility.",
      "votes": null
    },
    {
      "id": "1573316",
      "postDate": "11/06/2021 14:09:39",
      "content": "<p><a href=\"https://www.kaggle.com/ayuraj\" target=\"_blank\">@ayuraj</a> sir, tagging some of the organizers and Kaggle team involved might help.</p>",
      "rawMarkdown": "ayuraj sir, tagging some of the organizers and Kaggle team involved might help.",
      "votes": null
    },
    {
      "id": "1573638",
      "postDate": "11/06/2021 19:37:11",
      "content": "<p>Done. Thanks.</p>",
      "rawMarkdown": "Done. Thanks.",
      "votes": null
    },
    {
      "id": "1590567",
      "postDate": "11/21/2021 12:35:26",
      "content": "<p>Hello Ayush, great question!</p>\n<blockquote>\n  <p>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees</p>\n</blockquote>\n<p>Would you please expand on this a little bit? In particular, what do you mean by different trade fees and how is the difference in the OHLC data obtained? Thank you a lot in advance.</p>",
      "rawMarkdown": "Hello Ayush, great question!\n\n>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees\n\nWould you please expand on this a little bit? In particular, what do you mean by different trade fees and how is the difference in the OHLC data obtained? Thank you a lot in advance.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1573316,
      "author_name": "soumya9977",
      "author_url": "",
      "post_date": "11/06/2021 14:09:39",
      "content": "<p><a href=\"https://www.kaggle.com/ayuraj\" target=\"_blank\">@ayuraj</a> sir, tagging some of the organizers and Kaggle team involved might help.</p>",
      "votes": null,
      "replies": [
        {
          "id": 1573638,
          "author_name": "ayuraj",
          "author_url": "",
          "post_date": "11/06/2021 19:37:11",
          "content": "<p>Done. Thanks.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1590567,
      "author_name": "ichernov",
      "author_url": "",
      "post_date": "11/21/2021 12:35:26",
      "content": "<p>Hello Ayush, great question!</p>\n<blockquote>\n  <p>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees</p>\n</blockquote>\n<p>Would you please expand on this a little bit? In particular, what do you mean by different trade fees and how is the difference in the OHLC data obtained? Thank you a lot in advance.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1572722": "I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees. I wonder if the competition host can share some insight about the dataset creation process?\n\nI am trying to build a system that can automatically pull in trade data after 2021-09-21 which is then properly version controlled and easily distributed with fellow Kagglers. Thought of doing this since public data is allowed to use. \n\nOne major bottleneck is the data creation process. Any help will be appreciated. If you all think this is not the right path to follow, will love to know your thoughts. \n\n### Note\n\nSince the competition's LB score is of no use, if the data creation process is made clear, we can build our own data and evaluate the model on it. I think it will help come up with a more useful solution. \n\nTagging @andrewscherer for visibility.",
    "1573316": "ayuraj sir, tagging some of the organizers and Kaggle team involved might help.",
    "1573638": "Done. Thanks.",
    "1590567": "Hello Ayush, great question!\n\n>I just noticed that different exchanges have different OHLC (open high low close) data and it's because of different trade fees\n\nWould you please expand on this a little bit? In particular, what do you mean by different trade fees and how is the difference in the OHLC data obtained? Thank you a lot in advance."
  },
  "source": "meta"
}