{
  "id": 475548,
  "title": "train_applprev_1_0",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/475548",
  "author_name": "",
  "post_date": "2024-02-08T20:14:40.624293Z",
  "votes": 15,
  "comment_count": 5,
  "views": 0,
  "content": "<p>I'm not sure my understanding of the relationship linked by num_group1 in the train_applprev_1_0 is correct.</p>\n<p>My understanding is that the train_applprev_1_0 is the internal \"application previous\" i.e. history file.</p>\n<p>For num_group1&gt;0 (up to 19) these are the previous loan applications made by <strong>the same person</strong>, hence the person's loan application history. </p>\n<p>It looks like num_group1 is in reverse chronological order where 0 represents the current application and 19 the oldest, then for a given case_id, different values of num_group1 would represent the same individual's loan application history over time.</p>\n<p>Is my understanding correct?</p>",
  "messages": [
    {
      "id": "2643418",
      "postDate": "02/08/2024 20:14:40",
      "content": "<p>I'm not sure my understanding of the relationship linked by num_group1 in the train_applprev_1_0 is correct.</p>\n<p>My understanding is that the train_applprev_1_0 is the internal \"application previous\" i.e. history file.</p>\n<p>For num_group1&gt;0 (up to 19) these are the previous loan applications made by <strong>the same person</strong>, hence the person's loan application history. </p>\n<p>It looks like num_group1 is in reverse chronological order where 0 represents the current application and 19 the oldest, then for a given case_id, different values of num_group1 would represent the same individual's loan application history over time.</p>\n<p>Is my understanding correct?</p>",
      "rawMarkdown": "I'm not sure my understanding of the relationship linked by num_group1 in the train_applprev_1_0 is correct.\n\nMy understanding is that the train_applprev_1_0 is the internal \"application previous\" i.e. history file.\n\nFor num_group1>0 (up to 19) these are the previous loan applications made by **the same person**, hence the person's loan application history. \n\nIt looks like num_group1 is in reverse chronological order where 0 represents the current application and 19 the oldest, then for a given case_id, different values of num_group1 would represent the same individual's loan application history over time.\n\nIs my understanding correct?",
      "votes": null
    },
    {
      "id": "2643933",
      "postDate": "02/09/2024 07:23:08",
      "content": "<blockquote>\n  <p>Is my understanding correct?</p>\n</blockquote>\n<p>Yes, that is exactly what it means in this specific data file.</p>",
      "rawMarkdown": ">Is my understanding correct?\n\nYes, that is exactly what it means in this specific data file.",
      "votes": null
    },
    {
      "id": "2643965",
      "postDate": "02/09/2024 07:47:16",
      "content": "<p>Thank you for your swift reply.</p>",
      "rawMarkdown": "Thank you for your swift reply.",
      "votes": null
    },
    {
      "id": "2653896",
      "postDate": "02/15/2024 17:38:51",
      "content": "<p>Hello </p>\n<p>I don't understand the relationship between the num_group's properly.  Take the data in files like train_credit_bureau_a_2_0 filtered for case 411 for example.  The row for July 2018 shows a 13 past due payment for 2, 17 (num_group1, num_group2) but on 1, 17  and 0, 17 there is no past due showing. How can this be? Shouldn't they all be the same?</p>",
      "rawMarkdown": "Hello \n\nI don't understand the relationship between the num_group's properly.  Take the data in files like train_credit_bureau_a_2_0 filtered for case 411 for example.  The row for July 2018 shows a 13 past due payment for 2, 17 (num_group1, num_group2) but on 1, 17  and 0, 17 there is no past due showing. How can this be? Shouldn't they all be the same?",
      "votes": null
    },
    {
      "id": "2654283",
      "postDate": "02/16/2024 03:16:33",
      "content": "<p>Some case_id have num_group1 values up to 12. <br>\nIf num_group1 represents the history of loan applications, then all values from 12 to 0 should be present. <br>\nHowever, there are cases where this is not true. Why is that?<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F15793987%2Fffa83cf74ac6148272dc2da211ce7e0a%2FScreenshot%202024-02-16%20at%2011.50.11AM.png?generation=1708053390383502&amp;alt=media\"></p>",
      "rawMarkdown": "Some case_id have num_group1 values up to 12. \nIf num_group1 represents the history of loan applications, then all values from 12 to 0 should be present. \nHowever, there are cases where this is not true. Why is that?![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F15793987%2Fffa83cf74ac6148272dc2da211ce7e0a%2FScreenshot%202024-02-16%20at%2011.50.11AM.png?generation=1708053390383502&alt=media)",
      "votes": null
    },
    {
      "id": "2777682",
      "postDate": "04/26/2024 19:07:12",
      "content": "<p>then what does same num_group1's mean?<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10672506%2F5096c1afb72145a95472f44484c71750%2FScreenshot%202024-04-27%20003356.png?generation=1714158422185327&amp;alt=media\"></p>",
      "rawMarkdown": "then what does same num_group1's mean?![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10672506%2F5096c1afb72145a95472f44484c71750%2FScreenshot%202024-04-27%20003356.png?generation=1714158422185327&alt=media)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2643933,
      "author_name": "jetakow",
      "author_url": "",
      "post_date": "02/09/2024 07:23:08",
      "content": "<blockquote>\n  <p>Is my understanding correct?</p>\n</blockquote>\n<p>Yes, that is exactly what it means in this specific data file.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2654283,
          "author_name": "busybee0828",
          "author_url": "",
          "post_date": "02/16/2024 03:16:33",
          "content": "<p>Some case_id have num_group1 values up to 12. <br>\nIf num_group1 represents the history of loan applications, then all values from 12 to 0 should be present. <br>\nHowever, there are cases where this is not true. Why is that?<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F15793987%2Fffa83cf74ac6148272dc2da211ce7e0a%2FScreenshot%202024-02-16%20at%2011.50.11AM.png?generation=1708053390383502&amp;alt=media\"></p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2643965,
      "author_name": "signator",
      "author_url": "",
      "post_date": "02/09/2024 07:47:16",
      "content": "<p>Thank you for your swift reply.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2653896,
      "author_name": "blakswan",
      "author_url": "",
      "post_date": "02/15/2024 17:38:51",
      "content": "<p>Hello </p>\n<p>I don't understand the relationship between the num_group's properly.  Take the data in files like train_credit_bureau_a_2_0 filtered for case 411 for example.  The row for July 2018 shows a 13 past due payment for 2, 17 (num_group1, num_group2) but on 1, 17  and 0, 17 there is no past due showing. How can this be? Shouldn't they all be the same?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2777682,
      "author_name": "ravikumarchavva",
      "author_url": "",
      "post_date": "04/26/2024 19:07:12",
      "content": "<p>then what does same num_group1's mean?<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10672506%2F5096c1afb72145a95472f44484c71750%2FScreenshot%202024-04-27%20003356.png?generation=1714158422185327&amp;alt=media\"></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2643418": "I'm not sure my understanding of the relationship linked by num_group1 in the train_applprev_1_0 is correct.\n\nMy understanding is that the train_applprev_1_0 is the internal \"application previous\" i.e. history file.\n\nFor num_group1>0 (up to 19) these are the previous loan applications made by **the same person**, hence the person's loan application history. \n\nIt looks like num_group1 is in reverse chronological order where 0 represents the current application and 19 the oldest, then for a given case_id, different values of num_group1 would represent the same individual's loan application history over time.\n\nIs my understanding correct?",
    "2643933": ">Is my understanding correct?\n\nYes, that is exactly what it means in this specific data file.",
    "2643965": "Thank you for your swift reply.",
    "2653896": "Hello \n\nI don't understand the relationship between the num_group's properly.  Take the data in files like train_credit_bureau_a_2_0 filtered for case 411 for example.  The row for July 2018 shows a 13 past due payment for 2, 17 (num_group1, num_group2) but on 1, 17  and 0, 17 there is no past due showing. How can this be? Shouldn't they all be the same?",
    "2654283": "Some case_id have num_group1 values up to 12. \nIf num_group1 represents the history of loan applications, then all values from 12 to 0 should be present. \nHowever, there are cases where this is not true. Why is that?![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F15793987%2Fffa83cf74ac6148272dc2da211ce7e0a%2FScreenshot%202024-02-16%20at%2011.50.11AM.png?generation=1708053390383502&alt=media)",
    "2777682": "then what does same num_group1's mean?![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10672506%2F5096c1afb72145a95472f44484c71750%2FScreenshot%202024-04-27%20003356.png?generation=1714158422185327&alt=media)"
  },
  "source": "meta"
}