{
  "id": 555034,
  "title": "Why are we not using time series type data prediction tools?",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/555034",
  "author_name": "",
  "post_date": "2025-01-04T19:11:13.222268700Z",
  "votes": null,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I had a doubt is i joined the competition i thought it about market prediction which is a time series data so we will use tools like fbprophet, ARIMA etc. But as of most of the public notebooks they have the, features to prediction approach. As per the dataset , can't the time_id and date_id be given actual time with gap and a starting date to predict the response as time-series ?</p>",
  "messages": [
    {
      "id": "3088533",
      "postDate": "01/04/2025 19:11:13",
      "content": "<p>I had a doubt is i joined the competition i thought it about market prediction which is a time series data so we will use tools like fbprophet, ARIMA etc. But as of most of the public notebooks they have the, features to prediction approach. As per the dataset , can't the time_id and date_id be given actual time with gap and a starting date to predict the response as time-series ?</p>",
      "rawMarkdown": "I had a doubt is i joined the competition i thought it about market prediction which is a time series data so we will use tools like fbprophet, ARIMA etc. But as of most of the public notebooks they have the, features to prediction approach. As per the dataset , can't the time_id and date_id be given actual time with gap and a starting date to predict the response as time-series ?",
      "votes": null
    },
    {
      "id": "3088616",
      "postDate": "01/04/2025 22:26:05",
      "content": "<p>I've spent a lot of time analyzing the responders, ARIMA won't work well here, and I've personally tried it with poor results. I'm gonna make a post about this in a little bit, but I believe I've reverse engineered what type of data the responders are, and what the tags mean in responders.csv.</p>",
      "rawMarkdown": "I've spent a lot of time analyzing the responders, ARIMA won't work well here, and I've personally tried it with poor results. I'm gonna make a post about this in a little bit, but I believe I've reverse engineered what type of data the responders are, and what the tags mean in responders.csv.",
      "votes": null
    },
    {
      "id": "3090627",
      "postDate": "01/07/2025 14:01:07",
      "content": "<p>Looking forward to your post.</p>",
      "rawMarkdown": "Looking forward to your post.",
      "votes": null
    },
    {
      "id": "3090962",
      "postDate": "01/07/2025 20:45:09",
      "content": "<p>Hmm any hints at what you have found? I spent a little bit of time investigating those tags, for features too, but didn't get any solid insights</p>",
      "rawMarkdown": "Hmm any hints at what you have found? I spent a little bit of time investigating those tags, for features too, but didn't get any solid insights",
      "votes": null
    },
    {
      "id": "3091265",
      "postDate": "01/08/2025 08:38:38",
      "content": "<p><a href=\"https://www.kaggle.com/competitions/jane-street-real-time-market-data-forecasting/discussion/555562\" target=\"_blank\">Reverse Engineering the Responders</a></p>",
      "rawMarkdown": "[Reverse Engineering the Responders](https://www.kaggle.com/competitions/jane-street-real-time-market-data-forecasting/discussion/555562)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3088616,
      "author_name": "johnpayne0",
      "author_url": "",
      "post_date": "01/04/2025 22:26:05",
      "content": "<p>I've spent a lot of time analyzing the responders, ARIMA won't work well here, and I've personally tried it with poor results. I'm gonna make a post about this in a little bit, but I believe I've reverse engineered what type of data the responders are, and what the tags mean in responders.csv.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3090627,
          "author_name": "michaeldelamaza",
          "author_url": "",
          "post_date": "01/07/2025 14:01:07",
          "content": "<p>Looking forward to your post.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 3090962,
          "author_name": "julianmukaj",
          "author_url": "",
          "post_date": "01/07/2025 20:45:09",
          "content": "<p>Hmm any hints at what you have found? I spent a little bit of time investigating those tags, for features too, but didn't get any solid insights</p>",
          "votes": null,
          "replies": [
            {
              "id": 3091265,
              "author_name": "tinkei",
              "author_url": "",
              "post_date": "01/08/2025 08:38:38",
              "content": "<p><a href=\"https://www.kaggle.com/competitions/jane-street-real-time-market-data-forecasting/discussion/555562\" target=\"_blank\">Reverse Engineering the Responders</a></p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3088533": "I had a doubt is i joined the competition i thought it about market prediction which is a time series data so we will use tools like fbprophet, ARIMA etc. But as of most of the public notebooks they have the, features to prediction approach. As per the dataset , can't the time_id and date_id be given actual time with gap and a starting date to predict the response as time-series ?",
    "3088616": "I've spent a lot of time analyzing the responders, ARIMA won't work well here, and I've personally tried it with poor results. I'm gonna make a post about this in a little bit, but I believe I've reverse engineered what type of data the responders are, and what the tags mean in responders.csv.",
    "3090627": "Looking forward to your post.",
    "3090962": "Hmm any hints at what you have found? I spent a little bit of time investigating those tags, for features too, but didn't get any solid insights",
    "3091265": "[Reverse Engineering the Responders](https://www.kaggle.com/competitions/jane-street-real-time-market-data-forecasting/discussion/555562)"
  },
  "source": "meta"
}