How do others handle Time difference between broker and local?
5 replies
seaton
8 years ago #115031
How are others handling differences in timezones with generated strategies, especially with time sensitive indicators? Are you accounting for broker and local time in your strategies? if so how?
Thanks
mikeyc
8 years ago #136540
Hi Seaton,
Most of my strategies are time sensitive. I have handled it in this simple way.
The data I download using TickStory. This has the option to shift the Dukascopy data to New York time (EST + 7 hrs). This means that the New York close (5PM EST) coincides with midnight on MetaTrader. Why is this important?
http://www.learntotradethemarket.com/meta-trader-demo-account-sign-up-new-york-close-charts
TickStory also applies New York daylight saving shifts to the data.
All good. Now what about the broker?
Simple. I only ever trade with brokers that:
- Trade in line with New York close
- Shift daylight saving inline with New York
Eg.
http://www.icmarkets.com/client-area/faq/
What is IC Markets GMT Offset?
To align the daily chart chart candles with New York close (5pm ET) IC Markets server time and charts are GMT + 2 or GMT + 3 when daylight savings is in effect.
The table below outlines the dates on which the server time is changed to reflect daylight savings each year.
Year DST Commencement Date DST Ending Date
2013 10th March 3rd November
2014 9th March 2nd November
2015 8th March 1st November
Daylight saving time begins at 2 am on the second Sunday in March and ends at 2 am on the first Sunday in November.
This just makes it so simple. No confusion, or shifting when daylight changes. Align data to New York, choose broker aligned to New York.
Threshold
8 years ago #136551
I make my historical data the same timezone as my broker (NYC close/Sydney Open) and build from that. MT4 uses the broker’s time so my local time is meaningless.
seaton
7 years ago #137217
@threshold Thanks for the Reply,
This is what I’m currently doing, however I’d like too use Asirikuy data for my mining and backtesting for timed strategies as it’s +1/2 so not in line with any of my brokers, sorry for the late reply btw, have only just got around to refreshing my backtest data and revisited this.
Have you tried to convert Asirikuy M1 data ?
AC1962
7 years ago #137302
Hi Seaton
Not until after gaining a few months of SQ backtest experience did I realise the importance of making sure that my M1 and Tick data in SQ be set to my Broker’s server time, not my local time. I live in UK: GMT+0 & DST+1, but my UK Broker’s servers are in Cyprus: GMT+2 & DST+1. I only realised importance after using MT4-TDS to find that a final short-listed USDJPY H1 strategy, that worked well in SQ, gave zero trades in MT4-TDS. Reason being that the strategy only entered trades on Sundays, which is ok at local UK time as markets open 10pm Sunday, but not ok at Cyprus server time as markets open at midnight. I thought this may have been a 1-off, but exactly the same thing happened a week later after processing a different Fx pair (what are the odds [?], after SQ creating and excluding millions of strategies and after full backtest checks, I end up with 2x short-listed EA’â„¢s, each based on a different Fx pair, that only enter trades on Sundays). Further repeat tests found that ~25% of my previous EA results were impacted by this time-zone difference, though the impact was not generally so great. So I nowadays always use my broker’â„¢s server time in SQ.
AC1962
seaton
7 years ago #137630
Nice Article today by Daniel of Asirikuy.com here explaining how to convert using python.
http://mechanicalforex.com/2016/06/dealing-with-forex-data-time-zones-using-pandas.html
Viewing 5 replies - 1 through 5 (of 5 total)