{
  "id": 553961,
  "title": "Time steps has increased from 849 steps a day to 968 steps a day!",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/553961",
  "author_name": "Mohammad Naziruddin",
  "post_date": "2024-12-29T14:11:47.086000",
  "votes": -2,
  "comment_count": 3,
  "views": 0,
  "content": "<p>I did an analysis of count of timesteps per day for the entire period separately for each symbol_id (asset). I have listed the resulting data below and also the code to generate it. My observation is that most of the assets had initial timesteps of 849 per day, and later switched to 968. Few assets that started late do not have the data in this initial period (4,6,18,24,27,28,31,32,35 and 37).</p>\n<p>Would we be able to include the initial period of data in our analysis? I am planning to use GPU and LSTM where, as I understand, time steps should be should be uniform.</p>\n<p>Here is my notebook for this analysis: <a href=\"https://www.kaggle.com/code/nazir65/janestreettimestepanalysis\" target=\"_blank\">https://www.kaggle.com/code/nazir65/janestreettimestepanalysis</a></p>",
  "messages": [
    {
      "id": 3083461,
      "postDate": "2024-12-29T14:11:47.087Z",
      "content": "<p>I did an analysis of count of timesteps per day for the entire period separately for each symbol_id (asset). I have listed the resulting data below and also the code to generate it. My observation is that most of the assets had initial timesteps of 849 per day, and later switched to 968. Few assets that started late do not have the data in this initial period (4,6,18,24,27,28,31,32,35 and 37).</p>\n<p>Would we be able to include the initial period of data in our analysis? I am planning to use GPU and LSTM where, as I understand, time steps should be should be uniform.</p>\n<p>Here is my notebook for this analysis: <a href=\"https://www.kaggle.com/code/nazir65/janestreettimestepanalysis\" target=\"_blank\">https://www.kaggle.com/code/nazir65/janestreettimestepanalysis</a></p>",
      "rawMarkdown": "I did an analysis of count of timesteps per day for the entire period separately for each symbol_id (asset). I have listed the resulting data below and also the code to generate it. My observation is that most of the assets had initial timesteps of 849 per day, and later switched to 968. Few assets that started late do not have the data in this initial period (4,6,18,24,27,28,31,32,35 and 37).\n\nWould we be able to include the initial period of data in our analysis? I am planning to use GPU and LSTM where, as I understand, time steps should be should be uniform.\n\nHere is my notebook for this analysis: https://www.kaggle.com/code/nazir65/janestreettimestepanalysis",
      "votes": -2
    },
    {
      "id": 3084483,
      "postDate": "2024-12-30T22:04:52.673Z",
      "content": "<p>There are lots of missing symbols, especially for earlier <code>date_id</code> entries. I assume that they didn't monitor all symbols for the entire ~4 years of data generation, or chose to exclude them for some reason.</p>\n<p>Part of the difficulty in this competition is how to deal with missing symbols.</p>\n<p>There is quite a bit of discussion about this on the discussion forum.</p>",
      "rawMarkdown": "There are lots of missing symbols, especially for earlier `date_id` entries. I assume that they didn't monitor all symbols for the entire ~4 years of data generation, or chose to exclude them for some reason.\n\nPart of the difficulty in this competition is how to deal with missing symbols.\n\nThere is quite a bit of discussion about this on the discussion forum.",
      "replies": [
        {
          "id": 3091519,
          "postDate": "2025-01-08T13:39:27.370Z",
          "content": "<p>ok. will go through the discussion forum.</p>",
          "rawMarkdown": "ok. will go through the discussion forum."
        }
      ]
    },
    {
      "id": 3083788,
      "postDate": "2024-12-30T03:08:29.130Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 3084483,
      "author_name": "seamus barnes",
      "author_url": "",
      "post_date": "2024-12-30T22:04:52.673000",
      "content": "<p>There are lots of missing symbols, especially for earlier <code>date_id</code> entries. I assume that they didn't monitor all symbols for the entire ~4 years of data generation, or chose to exclude them for some reason.</p>\n<p>Part of the difficulty in this competition is how to deal with missing symbols.</p>\n<p>There is quite a bit of discussion about this on the discussion forum.</p>",
      "votes": 0,
      "replies": [
        {
          "id": 3091519,
          "author_name": "Mohammad Naziruddin",
          "author_url": "",
          "post_date": "2025-01-08T13:39:27.370000",
          "content": "<p>ok. will go through the discussion forum.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 3083788,
      "author_name": "",
      "author_url": "",
      "post_date": "2024-12-30T03:08:29.130000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3083461": "I did an analysis of count of timesteps per day for the entire period separately for each symbol_id (asset). I have listed the resulting data below and also the code to generate it. My observation is that most of the assets had initial timesteps of 849 per day, and later switched to 968. Few assets that started late do not have the data in this initial period (4,6,18,24,27,28,31,32,35 and 37).\n\nWould we be able to include the initial period of data in our analysis? I am planning to use GPU and LSTM where, as I understand, time steps should be should be uniform.\n\nHere is my notebook for this analysis: https://www.kaggle.com/code/nazir65/janestreettimestepanalysis",
    "3084483": "There are lots of missing symbols, especially for earlier `date_id` entries. I assume that they didn't monitor all symbols for the entire ~4 years of data generation, or chose to exclude them for some reason.\n\nPart of the difficulty in this competition is how to deal with missing symbols.\n\nThere is quite a bit of discussion about this on the discussion forum.",
    "3083788": ""
  }
}