{
  "id": 115940,
  "title": "Is local time in meters readings adjusted for daylight saving?",
  "url": "/competitions/ashrae-energy-prediction/discussion/115940",
  "author_name": "",
  "post_date": "2019-11-06T07:00:52.344935700Z",
  "votes": 7,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I am trying to match weather data and meters readings properly, based on the guessed/identified Site locations. One problem is that apparently local time zones have DST (Dayllight saving) changes which causes non-existent or ambiguous readings at the time of DST change when localizing meter readings data.\nThis is an example for US/Eastern Timezone causing NonExistentTimeError:\n```\nts1 = '2016-03-13 02:00:00'\nts1 = pd.to_datetime (ts1)\nts1 = pd.Series(ts1)\nts1.dt.tz_localize ('US/Eastern')</p>\n\n<p>Output: NonExistentTimeError: 2016-03-13 02:00:00\n```\nSince meters data is continuous and has readings which apparently correspond to non-existing times, I suspect that those readings are not reflecting DST changes. Is that reasonable? <strong>Is this common practice not to make DST changes for utilities meters?</strong></p>",
  "messages": [
    {
      "id": "666514",
      "postDate": "11/06/2019 07:00:52",
      "content": "<p>I am trying to match weather data and meters readings properly, based on the guessed/identified Site locations. One problem is that apparently local time zones have DST (Dayllight saving) changes which causes non-existent or ambiguous readings at the time of DST change when localizing meter readings data.\nThis is an example for US/Eastern Timezone causing NonExistentTimeError:\n```\nts1 = '2016-03-13 02:00:00'\nts1 = pd.to_datetime (ts1)\nts1 = pd.Series(ts1)\nts1.dt.tz_localize ('US/Eastern')</p>\n\n<p>Output: NonExistentTimeError: 2016-03-13 02:00:00\n```\nSince meters data is continuous and has readings which apparently correspond to non-existing times, I suspect that those readings are not reflecting DST changes. Is that reasonable? <strong>Is this common practice not to make DST changes for utilities meters?</strong></p>",
      "rawMarkdown": "I am trying to match weather data and meters readings properly, based on the guessed/identified Site locations. One problem is that apparently local time zones have DST (Dayllight saving) changes which causes non-existent or ambiguous readings at the time of DST change when localizing meter readings data.\nThis is an example for US/Eastern Timezone causing NonExistentTimeError:\n```\nts1 = '2016-03-13 02:00:00'\nts1 = pd.to_datetime (ts1)\nts1 = pd.Series(ts1)\nts1.dt.tz_localize ('US/Eastern')\n\nOutput: NonExistentTimeError: 2016-03-13 02:00:00\n```\nSince meters data is continuous and has readings which apparently correspond to non-existing times, I suspect that those readings are not reflecting DST changes. Is that reasonable? **Is this common practice not to make DST changes for utilities meters?**",
      "votes": null
    },
    {
      "id": "676373",
      "postDate": "11/19/2019 05:59:01",
      "content": "<p>Hi,\nHow did you solve this in the end ? </p>",
      "rawMarkdown": "Hi,\nHow did you solve this in the end ?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 676373,
      "author_name": "alicegabriel",
      "author_url": "",
      "post_date": "11/19/2019 05:59:01",
      "content": "<p>Hi,\nHow did you solve this in the end ? </p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "666514": "I am trying to match weather data and meters readings properly, based on the guessed/identified Site locations. One problem is that apparently local time zones have DST (Dayllight saving) changes which causes non-existent or ambiguous readings at the time of DST change when localizing meter readings data.\nThis is an example for US/Eastern Timezone causing NonExistentTimeError:\n```\nts1 = '2016-03-13 02:00:00'\nts1 = pd.to_datetime (ts1)\nts1 = pd.Series(ts1)\nts1.dt.tz_localize ('US/Eastern')\n\nOutput: NonExistentTimeError: 2016-03-13 02:00:00\n```\nSince meters data is continuous and has readings which apparently correspond to non-existing times, I suspect that those readings are not reflecting DST changes. Is that reasonable? **Is this common practice not to make DST changes for utilities meters?**",
    "676373": "Hi,\nHow did you solve this in the end ?"
  },
  "source": "meta"
}