Adding Quandl's EOD

[mueljul]

Hi Josh, your program is awesome, thank you for making it available and maintaining it. I would like to add Quandl's EOD dataset as a source to complement the somewhat disappointing WIKI data. Though, I am running into an issue. Maybe you could tell what I am doing wrong and how to add Quandl's EOD? I am receiving this error message: (AAPL as an example, not working for any symbol, despite > 9000 tsids are recognised correctly) _"The data source for EOD/AAPL is not implemented in the price extractor. Please define the columns in QuandlDownload.download_quandldata."

So I guess something is wrong with class QuandlDownload(object) -> download_quandl_data or class QuandlDataExtraction(object) ?

I tried adding a paragraph elif q_code[:4] == 'EOD': in the download_quandl_data function and a elif self.download_selection[:4] == 'eod': in the init function of class QuandlDataExtraction(object). I have also added a section in the build_symbology.py to include EOD/ . But that's not doing the job. What am I missing here?

[camisatx]

Hey Julien,

Adding the EOD dataset is a fantastic suggestion. I really like how it will provide high quality unadjusted prices, splits and dividends. This is especially pertinent since the Yahoo Finance and Google Finance extractors are currently broken.

I quickly looked things over, and these are the changes that are required:

• Build quandl_eod symbols from the ticker csv file and link them with existing tsid symbols
• Add processing code in the QuandlDownload class, including data ingest code (for prices, splits and dividends)

This is kind of an extensive addition, but I really think it will be worth it. If you don't mind, I think I can add this functionality over the weekend.

[mueljul]

That would be great!

[camisatx]

I updated the system so Quandl EOD can now be downloaded. I only tested it with the few dozen test codes, but it appears to work fine.

Also, I added a price adjustment function for converting the raw prices into adjusted prices. Check out the notes I wrote here.

To actually download the EOD data:

• If using the script, run python pySecMaster/pySecMaster/pySecMaster.py --daily-downloads quandl.eod -v
• If using Docker, open the Dockerfile, comment out the second to last line (with quandl.wiki) and uncomment the last line (with quandl.eod).

You can download both WIKI and EOD if you want, by using --daily-download quandl.wiki quandl.eod.

I'm closing this issue, but feel free to ask any questions/bugs you have!