Importing Dukascopy Historical data into zorro

Hi everyone,

I would appreciate if anyone could explain me how to import historical data into zorro. I have historical data from Tick data downloader (strategy quant).

Thanks in advance!

I have put the csv file in the history folder and I have put the file directory:

string InName = "History\EURUSD_H1.csv"; // name of the CSV file


But it doesn't work coudl anyone help me? thanks in advance

Repilado: Hello, I described the process in this post: http://www.opserver.de/ubb7/ubbthreads.php?ubb=showflat&Number=459580#Post459580

I'm currently trying to update the python code and create a bash script for batch conversion into .t6 format. Hopefully I'll update the relevant thread with the new scripts soon.

Thanks a lot pcz I have read your post and I found it extremely useful. You use Strategy Quant to download data but it is not possible to download de bid and ask prices. It is due to the fact that I have the free version of Tickdata?

Thanks in advance pcz,

I've used only SQ Tick Downloader (which is free) so I have no experience with SQ itself but the trial should have all the features of the full version. The only difference is that you can use it only while connected to the Internet.

Yes I was wrong in the previous answer I was downloading the 1H data, not tick data. Yes of course downloading tick data you have the following format: YYYY-MM-DD hh:mm:ss.fff,bid,ask,bidVol,askVol.

Thanks again for your help!

If you sign up with Dukas for a free demo, their JForex client has a very powerful historical data manager which allows you to download the tick data over any period or turn it into bars in virtually any format you require.

An 8 minute candle or a 27 tick bar? No problem, if you have the need :-)

You can also select from a variety of date formats, file formats etc.

A great service to the forex community.

scotpip: I've noticed that when downloading data via Dukascopy web interface with GMT set, there are different times for tick (or S1) and M1 data downloaded before DST change (i.e. in the first part of the year). I asked on Dukascopy forum but received no answer. Have you ever checked? Is it the same for JForex client?

For example:
- AUD/NZD tick data exported for 3.10.2016 start at 23:00 (on 2.10.) with ask price 1.05245
- AUD/NZD M1 data exported for 3.10.2016 start at 00:00 (on 3.10.) with open price 1.05245

Thanks a lot scotpip for your help,

I have heard about it. Now I have been using it and it is right what pcz says about the web interface of Dukascopy. I don't know why it is happening, but I suppose that in JForex client you are going to have the same problem cause Dukascopy web interface is based on JForex. Correct me if I'm wrong.

Hi

First, I'd strongly recommend installing the Java JForex client - it's much more convenient than the web interface as you can select date ranges, as well as a number of different bar types, date formats, file formats etc. It only takes a couple of minutes to register.

Second, I don't use time based bars or ticks so I haven't come across the issue with DST. I understood that the downloaded data is normalised to GMT. If there is an issue, Dukas usually get around to answering on the forum but they can be a bit slow. Usually if you post something in the bug section it gets a quicker reply.

Please report back with anything you find out...

Is the data offered by dukascopy better than the data provided by Quandl or yahoo through zorro?

AFAIK Yahoo and Quandl don't provide free Forex tick data. And from my experience Dukascopy is superior to many other sources. But if you have the possibility to use your broker's data, it's always the best solution. Unfortunately many Forex brokers don't provide historical data. In that case you have to use data from different source and check the backtest results against data collected from your broker and / or live trading.

Thank you pcz!

Anyone knows how is the quality of historic data from FXCM or Oanda? Are them serious data providers as Dukascopy?

I know IB does not provide serious price history.

Thanks in advance.