{
  "id": 118670,
  "title": "Solving with Time Series ",
  "url": "/competitions/ashrae-energy-prediction/discussion/118670",
  "author_name": "Anjeet",
  "post_date": "2019-11-23T10:45:57.589000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hi All,</p>\n\n<p>I have been trying to analyse the problem and come up with one of the option to predict using  Time series . </p>\n\n<p>Looking for some advise based on my understanding.\nAnalysing the training data, we have  1400 + building id and with daily meter reading data. \nWhereas  meter reading data among building id seems not correlated or any relation, so can we consider only few building ids (randomly) and address prediction process separately for each building id data set.</p>\n\n<p>At the same time, these individual building ids's meter reading data showing trend and correlation. </p>\n\n<p>Please add some insight if any solving this problem using time series </p>\n\n<p>Will post the progress soon</p>\n\n<p>Thanks \nAnjeet Kumar</p>",
  "messages": [
    {
      "id": 679768,
      "postDate": "2019-11-23T10:45:57.590Z",
      "content": "<p>Hi All,</p>\n\n<p>I have been trying to analyse the problem and come up with one of the option to predict using  Time series . </p>\n\n<p>Looking for some advise based on my understanding.\nAnalysing the training data, we have  1400 + building id and with daily meter reading data. \nWhereas  meter reading data among building id seems not correlated or any relation, so can we consider only few building ids (randomly) and address prediction process separately for each building id data set.</p>\n\n<p>At the same time, these individual building ids's meter reading data showing trend and correlation. </p>\n\n<p>Please add some insight if any solving this problem using time series </p>\n\n<p>Will post the progress soon</p>\n\n<p>Thanks \nAnjeet Kumar</p>",
      "rawMarkdown": "Hi All,\n\nI have been trying to analyse the problem and come up with one of the option to predict using  Time series . \n\nLooking for some advise based on my understanding.\nAnalysing the training data, we have  1400 + building id and with daily meter reading data. \nWhereas  meter reading data among building id seems not correlated or any relation, so can we consider only few building ids (randomly) and address prediction process separately for each building id data set.\n\nAt the same time, these individual building ids's meter reading data showing trend and correlation. \n\nPlease add some insight if any solving this problem using time series \n\nWill post the progress soon\n\nThanks \nAnjeet Kumar\n"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "679768": "Hi All,\n\nI have been trying to analyse the problem and come up with one of the option to predict using  Time series . \n\nLooking for some advise based on my understanding.\nAnalysing the training data, we have  1400 + building id and with daily meter reading data. \nWhereas  meter reading data among building id seems not correlated or any relation, so can we consider only few building ids (randomly) and address prediction process separately for each building id data set.\n\nAt the same time, these individual building ids's meter reading data showing trend and correlation. \n\nPlease add some insight if any solving this problem using time series \n\nWill post the progress soon\n\nThanks \nAnjeet Kumar\n"
  }
}