{
  "id": 478631,
  "title": "What do the columns \"pmts_dpdvalue_108P\" and \"pmts_pmtsoverdue_635A\" of \"train_credit_bureau_b_2.csv\" really mean?",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/478631",
  "author_name": "",
  "post_date": "2024-02-21T15:53:17.878393200Z",
  "votes": 5,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I am having a hard time understanding what the columns<code>pmts_dpdvalue_108P</code> and <code>pmts_pmtsoverdue_635A</code> in <code>train_credit_bureau_b_2.csv</code> really mean. Their definitions as given by <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/data\" target=\"_blank\"><code>feature_definitions.csv</code></a> are a little bit ambiguous:</p>\n<hr>\n<table>\n<thead>\n<tr>\n<th>Variable</th>\n<th>Description</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td><code>pmts_dpdvalue_108P</code></td>\n<td>Value of past due payment for active contract (num_group1 - existing contract, num_group2 - payment)</td>\n</tr>\n<tr>\n<td><code>pmts_pmtsoverdue_635A</code></td>\n<td>Active contract that has overdue payments (num_group1 - existing contract, num_group2 - payment)</td>\n</tr>\n</tbody>\n</table>\n<hr>\n<ul>\n<li><p>Is <code>pmts_dpdvalue_108P</code> the equivalent number of past due days for past due payments?</p></li>\n<li><p>Is <code>pmts_pmts-overdue_635A</code> the number of past due payments? <strong>I noticed that the entries of this column are multiples of 0.2 (why?)</strong>.</p></li>\n</ul>",
  "messages": [
    {
      "id": "2661949",
      "postDate": "02/21/2024 15:53:17",
      "content": "<p>I am having a hard time understanding what the columns<code>pmts_dpdvalue_108P</code> and <code>pmts_pmtsoverdue_635A</code> in <code>train_credit_bureau_b_2.csv</code> really mean. Their definitions as given by <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/data\" target=\"_blank\"><code>feature_definitions.csv</code></a> are a little bit ambiguous:</p>\n<hr>\n<table>\n<thead>\n<tr>\n<th>Variable</th>\n<th>Description</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td><code>pmts_dpdvalue_108P</code></td>\n<td>Value of past due payment for active contract (num_group1 - existing contract, num_group2 - payment)</td>\n</tr>\n<tr>\n<td><code>pmts_pmtsoverdue_635A</code></td>\n<td>Active contract that has overdue payments (num_group1 - existing contract, num_group2 - payment)</td>\n</tr>\n</tbody>\n</table>\n<hr>\n<ul>\n<li><p>Is <code>pmts_dpdvalue_108P</code> the equivalent number of past due days for past due payments?</p></li>\n<li><p>Is <code>pmts_pmts-overdue_635A</code> the number of past due payments? <strong>I noticed that the entries of this column are multiples of 0.2 (why?)</strong>.</p></li>\n</ul>",
      "rawMarkdown": "I am having a hard time understanding what the columns`pmts_dpdvalue_108P` and `pmts_pmtsoverdue_635A` in `train_credit_bureau_b_2.csv` really mean. Their definitions as given by [`feature_definitions.csv`](https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/data) are a little bit ambiguous:\n\n---\n\n| Variable | Description |\n| --- | --- |\n| `pmts_dpdvalue_108P` | Value of past due payment for active contract (num_group1 - existing contract, num_group2 - payment) |\n| `pmts_pmtsoverdue_635A` | Active contract that has overdue payments (num_group1 - existing contract, num_group2 - payment) |\n\n---\n\n- Is `pmts_dpdvalue_108P` the equivalent number of past due days for past due payments?\n\n- Is `pmts_pmts-overdue_635A` the number of past due payments? **I noticed that the entries of this column are multiples of 0.2 (why?)**.",
      "votes": null
    },
    {
      "id": "2674295",
      "postDate": "02/29/2024 08:27:55",
      "content": "<p>I am confused too, do you understand the true meaning already?</p>",
      "rawMarkdown": "I am confused too, do you understand the true meaning already?",
      "votes": null
    },
    {
      "id": "2674505",
      "postDate": "02/29/2024 11:15:23",
      "content": "<p>I still don't. My best guesses are the ones that I presented in my question.</p>",
      "rawMarkdown": "I still don't. My best guesses are the ones that I presented in my question.",
      "votes": null
    },
    {
      "id": "2681643",
      "postDate": "03/04/2024 18:50:27",
      "content": "<p>Hey, is there any discussion about the schema of the dataset anywhere.</p>",
      "rawMarkdown": "Hey, is there any discussion about the schema of the dataset anywhere.",
      "votes": null
    },
    {
      "id": "2694606",
      "postDate": "03/13/2024 06:50:15",
      "content": "<p>The columns \"pmts_dpdvalue_108P\" and \"pmts_pmtsoverdue_635A\" are related to payment delinquency information in a credit card dataset.</p>\n<p>\"pmts_dpdvalue_108P\": This column likely represents the Days Past Due (DPD) for payments. DPD is a measure of how late a borrower is on their payment. For example, if a borrower misses a payment due on January 1st and pays on January 15th, the DPD would be 15. It indicates the number of days a payment is past its due date.</p>\n<p>\"pmts_pmtsoverdue_635A\": This column may represent the amount of overdue payments. It could be the total amount of payments that are overdue for a specific period. This information helps in understanding the financial health of the borrower and their repayment behavior.</p>\n<p>In summary:</p>\n<p>\"pmts_dpdvalue_108P\": Days Past Due for payments, indicating how late payments are.<br>\n\"pmts_pmtsoverdue_635A\": Amount of overdue payments, showing the total amount of payments that are overdue.</p>\n<h2>These columns are important for credit risk analysis as they provide insights into the borrower's payment behavior and whether they are falling behind on their payments.</h2>\n<p>Answers from the GPT :)</p>",
      "rawMarkdown": "The columns \"pmts_dpdvalue_108P\" and \"pmts_pmtsoverdue_635A\" are related to payment delinquency information in a credit card dataset.\n\n\"pmts_dpdvalue_108P\": This column likely represents the Days Past Due (DPD) for payments. DPD is a measure of how late a borrower is on their payment. For example, if a borrower misses a payment due on January 1st and pays on January 15th, the DPD would be 15. It indicates the number of days a payment is past its due date.\n\n\"pmts_pmtsoverdue_635A\": This column may represent the amount of overdue payments. It could be the total amount of payments that are overdue for a specific period. This information helps in understanding the financial health of the borrower and their repayment behavior.\n\nIn summary:\n\n\"pmts_dpdvalue_108P\": Days Past Due for payments, indicating how late payments are.\n\"pmts_pmtsoverdue_635A\": Amount of overdue payments, showing the total amount of payments that are overdue.\nThese columns are important for credit risk analysis as they provide insights into the borrower's payment behavior and whether they are falling behind on their payments.\n--------------\nAnswers from the GPT :)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2674295,
      "author_name": "thelma1223",
      "author_url": "",
      "post_date": "02/29/2024 08:27:55",
      "content": "<p>I am confused too, do you understand the true meaning already?</p>",
      "votes": null,
      "replies": [
        {
          "id": 2674505,
          "author_name": "eliocordeiropereira",
          "author_url": "",
          "post_date": "02/29/2024 11:15:23",
          "content": "<p>I still don't. My best guesses are the ones that I presented in my question.</p>",
          "votes": null,
          "replies": [
            {
              "id": 2681643,
              "author_name": "abhi3005",
              "author_url": "",
              "post_date": "03/04/2024 18:50:27",
              "content": "<p>Hey, is there any discussion about the schema of the dataset anywhere.</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    },
    {
      "id": 2694606,
      "author_name": "sani84",
      "author_url": "",
      "post_date": "03/13/2024 06:50:15",
      "content": "<p>The columns \"pmts_dpdvalue_108P\" and \"pmts_pmtsoverdue_635A\" are related to payment delinquency information in a credit card dataset.</p>\n<p>\"pmts_dpdvalue_108P\": This column likely represents the Days Past Due (DPD) for payments. DPD is a measure of how late a borrower is on their payment. For example, if a borrower misses a payment due on January 1st and pays on January 15th, the DPD would be 15. It indicates the number of days a payment is past its due date.</p>\n<p>\"pmts_pmtsoverdue_635A\": This column may represent the amount of overdue payments. It could be the total amount of payments that are overdue for a specific period. This information helps in understanding the financial health of the borrower and their repayment behavior.</p>\n<p>In summary:</p>\n<p>\"pmts_dpdvalue_108P\": Days Past Due for payments, indicating how late payments are.<br>\n\"pmts_pmtsoverdue_635A\": Amount of overdue payments, showing the total amount of payments that are overdue.</p>\n<h2>These columns are important for credit risk analysis as they provide insights into the borrower's payment behavior and whether they are falling behind on their payments.</h2>\n<p>Answers from the GPT :)</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2661949": "I am having a hard time understanding what the columns`pmts_dpdvalue_108P` and `pmts_pmtsoverdue_635A` in `train_credit_bureau_b_2.csv` really mean. Their definitions as given by [`feature_definitions.csv`](https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/data) are a little bit ambiguous:\n\n---\n\n| Variable | Description |\n| --- | --- |\n| `pmts_dpdvalue_108P` | Value of past due payment for active contract (num_group1 - existing contract, num_group2 - payment) |\n| `pmts_pmtsoverdue_635A` | Active contract that has overdue payments (num_group1 - existing contract, num_group2 - payment) |\n\n---\n\n- Is `pmts_dpdvalue_108P` the equivalent number of past due days for past due payments?\n\n- Is `pmts_pmts-overdue_635A` the number of past due payments? **I noticed that the entries of this column are multiples of 0.2 (why?)**.",
    "2674295": "I am confused too, do you understand the true meaning already?",
    "2674505": "I still don't. My best guesses are the ones that I presented in my question.",
    "2681643": "Hey, is there any discussion about the schema of the dataset anywhere.",
    "2694606": "The columns \"pmts_dpdvalue_108P\" and \"pmts_pmtsoverdue_635A\" are related to payment delinquency information in a credit card dataset.\n\n\"pmts_dpdvalue_108P\": This column likely represents the Days Past Due (DPD) for payments. DPD is a measure of how late a borrower is on their payment. For example, if a borrower misses a payment due on January 1st and pays on January 15th, the DPD would be 15. It indicates the number of days a payment is past its due date.\n\n\"pmts_pmtsoverdue_635A\": This column may represent the amount of overdue payments. It could be the total amount of payments that are overdue for a specific period. This information helps in understanding the financial health of the borrower and their repayment behavior.\n\nIn summary:\n\n\"pmts_dpdvalue_108P\": Days Past Due for payments, indicating how late payments are.\n\"pmts_pmtsoverdue_635A\": Amount of overdue payments, showing the total amount of payments that are overdue.\nThese columns are important for credit risk analysis as they provide insights into the borrower's payment behavior and whether they are falling behind on their payments.\n--------------\nAnswers from the GPT :)"
  },
  "source": "meta"
}