{
  "id": 536074,
  "title": "Is it a time series or regression problem?",
  "url": "/competitions/brist1d/discussion/536074",
  "author_name": "",
  "post_date": "2024-09-25T19:53:37.908282200Z",
  "votes": 3,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hi everyone, is it a time series or a simple regression problem?</p>",
  "messages": [
    {
      "id": "2998599",
      "postDate": "09/25/2024 19:53:37",
      "content": "<p>Hi everyone, is it a time series or a simple regression problem?</p>",
      "rawMarkdown": "Hi everyone, is it a time series or a simple regression problem?",
      "votes": null
    },
    {
      "id": "3002448",
      "postDate": "09/30/2024 04:21:21",
      "content": "<p>Seems more like a regression problem as thee is not much data per patient from what I see. But feel free to just test it out. :-) </p>",
      "rawMarkdown": "Seems more like a regression problem as thee is not much data per patient from what I see. But feel free to just test it out. :-)",
      "votes": null
    },
    {
      "id": "3008668",
      "postDate": "10/07/2024 00:16:41",
      "content": "<p>The VAR model didn't perform as expected; treating it as a simple regression model without adding the lags actually yielded better results</p>",
      "rawMarkdown": "The VAR model didn't perform as expected; treating it as a simple regression model without adding the lags actually yielded better results",
      "votes": null
    },
    {
      "id": "3013844",
      "postDate": "10/10/2024 15:38:00",
      "content": "<p>There is some data per patient, around 30 days I think. I tried to model it as a time series, also as an hourly sequence.<br>\nBut since it can simply seen as a table with a continous target, treating it as a regression problem yielded better results.</p>",
      "rawMarkdown": "There is some data per patient, around 30 days I think. I tried to model it as a time series, also as an hourly sequence.\nBut since it can simply seen as a table with a continous target, treating it as a regression problem yielded better results.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3002448,
      "author_name": "thomasmeiner",
      "author_url": "",
      "post_date": "09/30/2024 04:21:21",
      "content": "<p>Seems more like a regression problem as thee is not much data per patient from what I see. But feel free to just test it out. :-) </p>",
      "votes": null,
      "replies": [
        {
          "id": 3008668,
          "author_name": "yassinebenzekri",
          "author_url": "",
          "post_date": "10/07/2024 00:16:41",
          "content": "<p>The VAR model didn't perform as expected; treating it as a simple regression model without adding the lags actually yielded better results</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 3013844,
      "author_name": "raphaelvonlottner",
      "author_url": "",
      "post_date": "10/10/2024 15:38:00",
      "content": "<p>There is some data per patient, around 30 days I think. I tried to model it as a time series, also as an hourly sequence.<br>\nBut since it can simply seen as a table with a continous target, treating it as a regression problem yielded better results.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2998599": "Hi everyone, is it a time series or a simple regression problem?",
    "3002448": "Seems more like a regression problem as thee is not much data per patient from what I see. But feel free to just test it out. :-)",
    "3008668": "The VAR model didn't perform as expected; treating it as a simple regression model without adding the lags actually yielded better results",
    "3013844": "There is some data per patient, around 30 days I think. I tried to model it as a time series, also as an hourly sequence.\nBut since it can simply seen as a table with a continous target, treating it as a regression problem yielded better results."
  },
  "source": "meta"
}