{
  "id": 73239,
  "title": "Large range of purchase amount in historical transaction",
  "url": "/competitions/elo-merchant-category-recommendation/discussion/73239",
  "author_name": "",
  "post_date": "2018-12-01T01:11:57.383233200Z",
  "votes": 8,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi everyone,</p>\n\n<p>After exploring the historical transactions, I notice that there is one record with purchase amount &gt; 6,000,000 while majority of the data is negative. I was wondering if I should be concerned over this fact since the amount should have been normalised. </p>\n\n<p>Does anyone might have an explanation for this issue?</p>",
  "messages": [
    {
      "id": "430798",
      "postDate": "12/01/2018 01:11:57",
      "content": "<p>Hi everyone,</p>\n\n<p>After exploring the historical transactions, I notice that there is one record with purchase amount &gt; 6,000,000 while majority of the data is negative. I was wondering if I should be concerned over this fact since the amount should have been normalised. </p>\n\n<p>Does anyone might have an explanation for this issue?</p>",
      "rawMarkdown": "Hi everyone,\n\nAfter exploring the historical transactions, I notice that there is one record with purchase amount &gt; 6,000,000 while majority of the data is negative. I was wondering if I should be concerned over this fact since the amount should have been normalised. \n\nDoes anyone might have an explanation for this issue?",
      "votes": null
    },
    {
      "id": "434349",
      "postDate": "12/06/2018 08:45:09",
      "content": "<p>I have notice the same problem, <code>purchase amount</code> &gt; 6,000,000 maybe were caused by the inappropriate normalization function?</p>",
      "rawMarkdown": "I have notice the same problem, `purchase amount` &gt; 6,000,000 maybe were caused by the inappropriate normalization function?",
      "votes": null
    },
    {
      "id": "434773",
      "postDate": "12/06/2018 23:16:33",
      "content": "<p>Probably it is a person like this lady <a href=\"https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html\">https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html</a></p>\n\n<p>The trade amount mean  is close to zero. Having this huge single amount in the data set we get all amounts shifted below zero to balance this outlier.</p>",
      "rawMarkdown": "Probably it is a person like this lady https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html\n\nThe trade amount mean  is close to zero. Having this huge single amount in the data set we get all amounts shifted below zero to balance this outlier.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 434349,
      "author_name": "sunnymarkliu",
      "author_url": "",
      "post_date": "12/06/2018 08:45:09",
      "content": "<p>I have notice the same problem, <code>purchase amount</code> &gt; 6,000,000 maybe were caused by the inappropriate normalization function?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 434773,
      "author_name": "kmedvedev",
      "author_url": "",
      "post_date": "12/06/2018 23:16:33",
      "content": "<p>Probably it is a person like this lady <a href=\"https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html\">https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html</a></p>\n\n<p>The trade amount mean  is close to zero. Having this huge single amount in the data set we get all amounts shifted below zero to balance this outlier.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "430798": "Hi everyone,\n\nAfter exploring the historical transactions, I notice that there is one record with purchase amount &gt; 6,000,000 while majority of the data is negative. I was wondering if I should be concerned over this fact since the amount should have been normalised. \n\nDoes anyone might have an explanation for this issue?",
    "434349": "I have notice the same problem, `purchase amount` &gt; 6,000,000 maybe were caused by the inappropriate normalization function?",
    "434773": "Probably it is a person like this lady https://www.thestar.com/news/world/2018/10/10/woman-who-spent-21-million-at-harrods-fights-uk-order-to-explain-source-of-her-wealth.html\n\nThe trade amount mean  is close to zero. Having this huge single amount in the data set we get all amounts shifted below zero to balance this outlier."
  },
  "source": "meta"
}