Historical Market Data - Where can I get a better format?

Discussion in 'Trading Software' started by Jason-S, Jun 12, 2016.

  1. Jason-S

    Jason-S

    Hi,
    I'm new here and grateful this forum exists.

    Could you suggest some data resources that align better with what I'm trying to do?
    To begin I need to analyze 5 years of NASDAQ daily closing values of every stock on the exchange within 1 excel spreadsheet so that I can find the outliers among all of the symbols which meet my own preliminary set of criteria.

    I've attached a screenshot of the data layout I want. Check it out.
    Here's a written description.
    A vertical list of all symbols in column A. (starting A3).
    Each day (spanning 5 years) <---> One per column. (starting C3)
    Closing values entered for each symbol per day.
    *A bonus would be if I could have just one column listing the average daily volume over the 5 years. (starting B3). Not a deal breaker if I can't get this too.

    I bought 5 years of NASDAQ data from eoddata.com however the way the data is arranged in the files is far off from the layout I need to do my analysis, and so if you could suggest some data resources or solutions that could get me the arrangement of of that much data in excel in the way I've outlined (or close to it), that would be wonderful.

    Thank you for reading my post and I -sincerely- appreciate your help and direction.

    Jason
     
  2. Excel sucks for data science... Microsoft has now joined with the the R programming community... second... EOD data is cheap from anywhere... Reuters cqg etc etc
     
    mjl13 likes this.
  3. just21

    just21

  4. Or you could use R and quandl as I have .. excel sucks you will quickly reach its limitations
     
  5. userque

    userque

    I understand they already have the data.

    They'll likely need some VBA code to do this with less labor and in less time.

    I can't know if what they are doing will choke Excel; because they haven't said exactly what they'll be doing. Nor have they said how powerful the computer is they'll be running this on.

    If they can't code for themselves (likely the case); they can try sites like freelancer.com to hire someone to code it for them.
     
  6. Excel will bind up and hang on very small data sets... good luck...
     
  7. If there is a lot of data I would go with SQL:
    https://www.microsoft.com/en-us/download/details.aspx?id=42299
    1. it is FREE
    2. IT is fast and handle well big sets of data
    3. IF you get handy with SQL queries (very easy language) you may set different views for your needs
    4. If you not familiar with SQL queries, you will find huge online help in different communities, even here.
     
  8. Saving Time series data in a relational dB does not have any benefit..
     
    mjl13 likes this.
  9. mjl13

    mjl13

    R alone sucks for large datasets too. Try SAS or Python or an actual database to store your data.
     
  10. Completely false... I deal with hundreds of millions of rows of tick data very quickly with no dB... there's packages to parrellize operations as well.... just a statement in ignorance
     
    #10     Jun 13, 2016
    Oysteryx likes this.